Physical, Morphological and Chemical Structure & Property

Physical, Morphological and Chemical

Structure & Property Relationships for

α-Keratins in Bleached Human Hair

A thesis submitted to The University of Manchester for

the degree of Doctor of Philosophy (PhD) in the Faculty

of Engineering and Physical Sciences

2013

Daijiazi Zhang

School of Materials

Table of Contents

2

Table of Contents

Abstract ......................................................................................................................... 14

Declaration .................................................................................................................... 15

Copyright Statement ..................................................................................................... 16

Acknowledgement......................................................................................................... 17

Abbreviations ................................................................................................................ 18

Chapter 1 Introduction ................................................................................................ 20

1.1 Morphology and Chemical Structure of Human Hair .................................... 20

1.1.1 Cuticle ..................................................................................................... 21

1.1.2 Cell Membrane Complex (CMC) ........................................................... 23

1.1.3 Cortex ...................................................................................................... 24

1.1.4 Medulla ................................................................................................... 26

1.2 The Chemical Composition of Human Hair Fibres ........................................ 27

1.2.1 Proteins .................................................................................................... 27

1.2.2 Water ........................................................................................................ 29

1.2.3 Lipids ....................................................................................................... 30

1.2.4 Pigments................................................................................................... 31

1.2.5 Chemical Elements .................................................................................. 32

1.3 Bonding Mechanisms in Keratin ...................................................................... 32

1.4 The Keratin-Water System ............................................................................... 34

1.5 Chemical Treatments ........................................................................................ 35

1.5.1 Bleaching Principle .................................................................................. 35

1.6 Objectives ......................................................................................................... 38

1.7 Reference .......................................................................................................... 39

Chapter 2 Sample Preparation .................................................................................... 43

2.1 Sample Material ............................................................................................... 43

2.2 Bleaching Products & Procedures .................................................................... 43

2.2.1 Bleaching Products .................................................................................. 43

2.2.2 Bleaching Procedure ................................................................................ 44

Chapter 3 SEM Investigation of the Surface Properties of Human Hair .................... 46

3.1 Introduction ...................................................................................................... 46

Table of Contents

3

3.1.1 Evaluation Method for Surface Properties............................................... 46

3.1.2 The SEM Investigation of Human Hair ................................................... 48

3.2 Objectives ........................................................................................................ 48

3.3 Experimental ................................................................................................... 49

3.4 Results and Discussions .................................................................................. 50

3.4.1 Morphological Structures of Bleached Hairs ........................................... 50

3.4.2 Knot Test.................................................................................................. 53

3.5 Conclusion ....................................................................................................... 63

3.6 Reference ......................................................................................................... 64

Chapter 4 Colour Measurement of Human Hair ....................................................... 66

4.1 Introduction ..................................................................................................... 66

4.1.1 Human Colour Vision & CIE Colour Space ............................................. 66

4.1.2 Colour Measurement................................................................................ 74

4.2 Objectives ........................................................................................................ 76

4.3 Experimental ................................................................................................... 76

4.4 Results and Discussion .................................................................................... 77

4.5 Conclusion ....................................................................................................... 90

4.6 References ....................................................................................................... 90

Chapter 5 Thermal Analysis of Human Hair ............................................................. 93

5.1 Thermal Analysis ............................................................................................ 93

5.1.1 Differential Scanning Calorimetry (DSC) ................................................. 93

5.1.2 Thermal Stability of Polymers ................................................................... 94

5.1.3 DSC Investigation of Keratins ................................................................... 99

5.1.4 Nonisothermal Kinetics of Keratin Thermal Denaturation..................... 103

5.2 Objectives ...................................................................................................... 110

5.3 Experimental ................................................................................................. 110

5.4 Results and Discussion ................................................................................. 112

5.4.1 DSC of Virgin Hair in Water .................................................................. 112

5.4.2 Water Absorption Effect on the Keratin ................................................. 115

5.4.3 DSC on Various Bleaches Effects .......................................................... 116

5.4.4 Nonisothermal Kinetics of α-keratin Thermal Denaturation .................. 121

5.4.5 DSC Deconvolution ................................................................................ 134

5.5 Morphological Changes on Three Ethnics Hair Types after DSC investigations

................................................................................................................................. 148

Table of Contents

4

5.5.1 Theory and Background.......................................................................... 148

5.5.2 Experimental ........................................................................................... 149

5.5.3 Results and Discussion ........................................................................... 151

5.6 Conclusion ..................................................................................................... 153

5.7 References ..................................................................................................... 154

Chapter 6 FTIR Investigations of Human Hair ........................................................ 163

6.1 Introduction .................................................................................................... 163

6.1.1 Infrared Spectrometer .............................................................................. 165

6.1.2 FTIR Sampling Techniques ..................................................................... 168

6.1.3 FTIR Technique and Protein Structure .................................................... 175

6.2 Objectives ....................................................................................................... 179

6.3 Experimental Work ........................................................................................ 180

6.3.1 Sample Preparation ................................................................................. 180

6.3.2 Spectral Processing ................................................................................. 182

6.4 Result and Discussion .................................................................................... 184

6.4.1 FTIR spectrum of Human Hair ............................................................... 184

6.4.2 Comparison of FTIR Spectra of Untreated and Bleached Human Hair . 189

6.5 Conclusion ...................................................................................................... 203

6.6 References ...................................................................................................... 205

Chapter 7 Summary and Conclusion ...................................................................... 210

7.1 Comparison of FTIR-KBr and DSC Results .................................................. 213

7.2 Future Work ................................................................................................... 214

Chapter 8 Appendix ................................................................................................ 215

Word Count: 54,024

List of Figures

5

List of Figures

Figure 1.1: Schematic diagram of a human hair with its various layers of cellular

structure ......................................................................................................................... 21

Figure 1.2: The fine structure of a human hair cuticle cell, which is separated from

the adjacent cell by the cell membrane complex .......................................................... 22

Figure 1.3: Schematic diagram of the rod domain of an intermediate filament dimer25

Figure 1.4: Higher order -keratin structure ................................................................ 26

Figure 1.5: The condensation reaction of amino acids .............................................. 27

Figure 1.6: Chemical structures of eumelanins and pheomelanins ............................. 31

Figure 1.7: Structural formula of a fictive peptide chain to illustrate five important

interactions between amino acid side-chains in keratins. ............................................. 33

Figure 1.8: Typical hydrogen bonds involving the atoms of oxygen and nitrogen .... 33

Figure 1.9: Cylindrical two-phase model of an -keratin fiber consisting of rigid

water-impenetrable phase C set parallel to the fiber axis, embedded in the matrix phase

M, which is water-penetrable ........................................................................................ 34

Figure 1.10: Scheme of the S-S cleavage mechanism for the bleaching process ........ 37

Figure 3.1: (A) Micrograph of an untreated hair fiber with somewhat broken edges

(B) Micrograph of a bleached hair fiber. The appearance is somewhat similar to

untreated hair, though slightly greater uplift of scales is noted. ................................... 50

Figure 3.2: (A) Micrograph of a bleached hair fiber, showing severe cuticle uplift

(B) Micrograph of a bleached hair fiber, the greater lift up of scales is noted, and the

hole in the cuticle is observed. ...................................................................................... 51

Figure 3.3: (A) Micrograph of bleached hair fiber showing exposed cortex

(B) Micrograph of bleached hair fiber, indicating a major lack of cuticle scales on the

fiber, with cuticle scales remnants remain. ................................................................... 51

Figure 3.4: (A) The fracture has exposed the underlying cortex (B) Macrofibrils of the

cortex are observed through the opening of the cuticle. ............................................... 52

Figure 3.5: Cortical macro-fibrils are clearly visible in the fracture zone of a bleached

hair fiber ........................................................................................................................ 52

Figure 3.6: Standard damage classification for mildly damaged fibers ....................... 54

Figure 3.7: Damage classification for heavily damaged fibers .................................... 55

Figure 3.8: Fiber count of virgin hair samples (N=30) in each classification ............. 56

List of Figures

6

Figure 3.9: Fiber count of 6% H202 bleached hair samples (N=30) in each

classification, at different bleaching times .................................................................... 58

Figure 3.10: Fibre count of 9% commercial bleached hair samples (N=30) in each

classification, at different bleaching times .................................................................... 59

Figure 3.11: Fiber count of commercial persulphate bleached hair samples (N=30) in

each classification, at different bleaching times ........................................................... 60

Figure 4.1: The visible portion of the spectrum is expanded to show the hues

associated with different wavelengths of light .............................................................. 66

Figure 4.2: Normalized response spectra of human eye cones, S, M, and L types, to

monochromatic spectral stimuli with wavelength in nanometres. ................................ 67

Figure 4.3: CIE-recommended illuminating and viewing geometries ........................ 71

Figure 4.4: Comparison between D/8 specular-included and -excluded modes ......... 71

Figure 4.5: CIE L*a*b* colour system ....................................................................... 73

Figure 4.6: Schematic of a reflection spectrophotometer with an integrating sphere . 75

Figure 4.7: The basic features of a colorimeter with 45/0 geometry .......................... 76

Figure 4.8: Values of a* for 3 bleaches: (6% H2O2, 9% commercial bleach,

commercial persulphate bleach) for up to 4 hours bleaching time ............................... 82

Figure 4.9: Values of b* for 3 bleaches: (6% H2O2, 9% commercial bleach,


Figure 4.10: Values of L* for 3 bleaches: (6% H2O2, 9% commercial bleach,


Figure 4.11: Values of ∆E for 3 bleaches: (6% H2O2, 9% commercial bleach,


Figure 4.12: The relationship of L*- and b*- values for 6% bleach over the 4h

bleaching time ............................................................................................................... 85

Figure 4.13: The relationship of L*- and b*- values for 9% commercial bleach over

the 4h bleaching time .................................................................................................... 86

Figure 4.14: The relationship of L*- and b*- values for commercial persulphate bleach

over the 4h bleaching time ............................................................................................ 86

Figure 4.15: The plot of L* vs. b* for hairs subjected to three types of bleaches ...... 87

Figure 4.16: Reflectance of 6% bleached hair samples in the visible spectrum ......... 88

Figure 4.17: Reflectance of Commercial persulphate bleached hair samples in the

visible spectrum ............................................................................................................ 89

List of Figures

7

Figure 4.18: Reflectance of 9% commercial bleached hair samples in the visible

spectrum ........................................................................................................................ 89

Figure 5.1: Schematic representation of the experimental setup in power compensation

DSC and heat flux DSC . ............................................................................................. 94

Figure 5.2: (A) describes the hydrogen bonds of a simple α-helix secondary structure

within one protein. (B) describes the energy states of protein denaturation ................. 96

Figure 5.3: Temperature-modulated DSC heating profile ........................................ 105

Figure 5.4: Typical non-reversing CpNR

DSC curve for untreated hair in water. ....... 113

Figure 5.5:Typical reversing CpR and non-reversing Cp

NR DSC curves for untreated

human hair in water ..................................................................................................... 114

Figure 5.6: Relative denaturation temperature TD R for 3 bleaches: (6% H2O2, 9%

Comercial Bleach, Commercial Persulphate bleach) for 3 hours bleaching time....... 119

Figure 5.7: Relative denaturation enthalpy ∆HDR for 3 bleaches: (6% H2O2, 9%

Commercial Bleach, commercial Persulphate bleach) for 3 hours bleaching time. ... 119

Figure 5.8: Plot of ∆HDR

against TDR for the 3 bleaches: (6% H2O2, 9% commercial

H202, commercial persulphate bleach) for up to 3 hours bleaching time. ................... 121

Figure 5.9: Denaturation non-reversing CpNR

DSC curves for virgin hair at various

heating rates ................................................................................................................ 123

Figure 5.10:Denaturation non-reversing CpNR

DSC curves for 2h commercially

bleached hair at various heating rates ......................................................................... 123

Figure 5.11: The relationship between heating rate, β (as lnβ) and 1/TD for virgin and

commercial persulphate bleached hair (2h) ................................................................ 124

Figure 5.12: Relationship of denaturation enthalpy ∆HD and lnβ for virgin and

bleached hair ............................................................................................................... 125


DSC curve for untreated mohair hair in water,

when the heating rate is 1oC/min ................................................................................ 127



when the heating rate is 3oC/min ................................................................................ 127

Figure 5.15: The relationship between heating rate, β (as lnβ) and 1/TD for mohair.128

Figure 5.16: Relationship of denaturation enthalpy ∆HD and lnβ for mohair ........... 129

Figure 5.17: Relationship of denaturation enthalpy ∆HD and lnβ for mohair and virgin

human hair ................................................................................................................... 130

Figure 5.18: Virgin human hair fiber snippets by SEM after being heated in the DSC

at various heating rates up to 7oc/min. ........................................................................ 131

List of Figures

8

Figure 5.19: Commercial persulphate bleached human hair fibre snippets (2h) after

SEM after being heated in the DSC at various heating rates up to 7oc/min................ 133

Figure 5.20: The DSC baseline of virgin human hair between the two chosen points, at

a heating rate of 3oC/min............................................................................................. 135

Figure 5.21: The corrected DSC peak of virgin human hair, at a heating rate of

3oC/min ....................................................................................................................... 136

Figure 5.22: Deconvolution of the MDSC-curve for virgin human hair into three

Gaussian distributions. The heating rate is 3oC/min. .................................................. 136

Figure 5.23: Denaturation temperatures of para-, ortho- cortical and background cell

types on the basis of three Gaussian distributions for 6% H202 and commercial

persulphate bleached hairs up to 3hr bleaching time. ................................................. 140


types on the basis of three Gaussian distributions for 9% commercial bleached hairs up

to 3hr bleaching time. .................................................................................................. 141

Figure 5.25: Fractional enthalpies of para-, ortho- cortical and background cell types

on the basis of three Gaussian distributions for 6% H202 bleached hairs up to 3hr

bleaching time. ............................................................................................................ 142


on the basis of three Gaussian distributions for 9% commercial bleached hairs up to

3hr bleaching time. ...................................................................................................... 142


on the basis of three Gaussian distributions for commercial persulphate bleached hairs

up to 3hr bleaching time. ............................................................................................. 143


types on the basis of three Gaussian distributions for virgin and commercial

persulphate bleached hairs (2h) at various heating rates. ............................................ 145


on the basis of three Gaussian distributions for virgin and commercial persulphate

bleached hairs (2h) at various heating rates. ............................................................... 146

Figure 5.30: Denaturation temperature for para-, ortho- and meso- cortical cell

fraction on the basis of three Gaussian distributions for virgin mohair at various

heating rates. ............................................................................................................... 147

Figure 5.31: Fractional enthalpies for para-, ortho- and meso-cortical cell fraction on

the basis of three Gaussian distributions for virgin mohair at various heating rates. . 148

List of Figures

9


DSC curve for three types of human hair in

water at a heating rate of 5oC/min ............................................................................... 150

Figure 5.33: SEM images of three types of hair after being heated to their denaturation

temperature and left at various isothermal times, before cooling ............................... 153

Figure 6.1: Schematic diagram of an FTIR spectrometer, showing its layout........... 166

Figure 6.2: An illustration of transmission FTIR sampling, where the infrared beam

passes through the sample and then strikes the detector ............................................. 168

Figure 6.3: Schematic diagram of a single reflection ATR showing the radiation path

through an infra-red transmitting crystal of high refractive index .............................. 174

Figure 6.4: The structure of an amino acid: the basic building block for making

proteins ........................................................................................................................ 175

Figure 6.5: The vibrations mainly responsible for the Amide I and Amide II bands in

the infrared spectra of proteins and polypeptides ....................................................... 176

Figure 6.6: The structures of the most important sulphur-containing amino acids in

weathered and bleached hair ....................................................................................... 177

Figure 6.7: KBr pellets of virgin and bleached hair and a pure KBr crystal ............ 182

Figure 6.8: The raw FTIR-ATR spectrum of an untreated hair sample ..................... 186

Figure 6.9: The raw FTIR-KBr spectrum of an untreated hair sample ...................... 186

Figure 6.10: Second derivative analysis of untreated hair (Batch I, 8cm) in the ATR

mode in the region 1240 - 980 cm-1

............................................................................ 188

Figure 6.11: Second derivative analysis of untreated hair (Batch I) in the transmission


............................................................................ 188

Figure 6.12: FTIR-ATR spectra of untreated and 3 types of 4h treated fibers at 1300 –

900cm-1

, after normalization ....................................................................................... 190

Figure 6.13: FTIR-ATR spectrum 2nd

derivative of untreated and 3 types of 4h treated

fibers at 1300 – 900cm-1

.............................................................................................. 191

Figure 6.14: Cysteic acid absolute valley depth for bleached hair measured using

FTIR-ATR after bleaching time of 0.5h ..................................................................... 195


FTIR-ATR after bleaching time of 2h ........................................................................ 196


FTIR-ATR after bleaching time of 4h. ....................................................................... 196

Figure 6.17: Oxidation effects on hairs measured using FTIR-ATR at various

oxidizing agents. ......................................................................................................... 198

List of Figures

10

Figure 6.18: Oxidation effects on bleached hair measured using FTIR-ATR at various

oxidizing agents .......................................................................................................... 199

Figure 6.19: FTIR-KBr spectrum of untreated and 3 types of 4h treated fibers at 1300

– 900cm-1

after normalization ..................................................................................... 200

Figure 6.21: Oxidation effects on bleached hair measured using FTIR-KBr for various


Figure 6.22: Oxidation effects on bleached hair measured using FTIR-KBr at various


Figure 6.23: Oxidation effects on bleached hair measured using FTIR-ATR and FTIR-

transmission mode for various oxidizing agents ......................................................... 204

Figure 7.1: Comparison of the relative cysteic acid absolute valley depth from FTIR-

KBr, with relative denaturation temperature from DSC measurements. .................... 214

List of Tables

11

List of Tables

Table 1.1:The various layers of the cuticle, their cystine content and details about their

properties ....................................................................................................................... 23

Table 1.2: Typical amino acid composition of human hair.......................................... 28

Table 3.1: Fibre count of virgin hair samples (N=30) in each classification ............... 56

Table 3.2: Fibre count of 6% H202 bleached hair samples ........................................... 57

Table 3.3: Fibre count of 9% commercial bleached hair samples ............................... 58

Table 3.4: Fibre count of commercial persulphate bleached hair samples .................. 60

Table 3.5: Overall damage classification of two groups of damage classes for virgin

hair samples ................................................................................................................... 61

Table 3.6: Overall damage classification of two groups of damage classes for 6% H202

bleached hair samples ................................................................................................... 62

Table 3.7: Overall damage classification of two damage types for 9% bleached hair

samples .......................................................................................................................... 62

Table 3.8:Overall damage classification of two damage types for commercial

persulphate bleached hair samples ................................................................................ 63

Table 4.1: The CIELAB results for 6% H2O2 bleached hair samples at different

bleaching times.............................................................................................................. 78

Table 4.2: The CIELAB results for 9% commercial bleached hair samples at different

bleaching times.............................................................................................................. 78

Table 4.3: The CIELAB results for commercial persulphate bleached hair samples at

different bleaching times ............................................................................................... 79

Table 4.4: The colour changes of ∆L*, ∆a*, ∆b* and ∆E* for 6% H2O2 bleached hair

samples at different bleaching times ............................................................................. 79

Table 4.5: The colour changes of ∆L*, ∆a*, ∆b* and ∆E* for 9% commercial

bleached hair samples at different bleaching times....................................................... 80

Table 4.6: The colour changes of ∆L*, ∆a*, ∆b* and ∆E* for commercial persulphate

bleached hair samples at different bleaching times....................................................... 80

Table 5.1: The parameters for the heating rate and modulation ................................ 111

List of Tables

12

Table 5.2: The comparison between denaturation temperatures TD and enthalpies ∆HD

for commercial persulphate bleached hairs before and after soaking in water for 48

hours ............................................................................................................................ 115

Table 5.3: The results for TD and ∆HD for the 6% bleached samples, in the form of the

arithmetic means ......................................................................................................... 117

Table 5.4: The results for TD and ∆HD for the 9% commercial bleached samples, in the

form of the arithmetic means ...................................................................................... 118

Table 5.5: The results for TD and ∆HD for the commercial persulphate bleached

samples, in the form of the arithmetic means ............................................................. 118

Table 5.6: Results for virgin hair at various heating rates ........................................ 122

Table 5.7: Results for commercial bleached human Hair (2h) at various heating rates

..................................................................................................................................... 122

Table 5.8: Results of mohair at various heating rates ............................................... 128

Table 5.9: Peak temperatures and fractional enthalpies for para- and ortho- cortical

cells and background for 6% H202 bleached hairs at various bleaching times ........... 138


cells and background for 9% commercial bleached hairs at various bleaching times.139


cells and background for commercial persulphate bleached hairs at various bleaching

times ............................................................................................................................ 139

Table 5.12: Peak temperatures and fractional enthalpies of three cell types in virgin

hair at various heating rates......................................................................................... 144

Table 5.13: Peak temperatures and fractional enthalpies of three cell types in

commercial persulphate bleached hair (2h) at various heating rates .......................... 144

Table 5.14: Peak temperatures and fractional enthalpies of para-, ortho- and meso-

cortical cells in virgin mohair for various heating rates .............................................. 147

Table 5.15: Description of SEM images for various hair samples............................. 151

Table 6.1: Values of the three IR regions ................................................................ 164

Table 6.2: Absolute valley depth for cysteic acid of five measurement points of virgin

hair for the second derivative ...................................................................................... 192

Table 6.3: Absolute valley depth for cysteic acid of five measurement points of three

bleached hairs for the second derivative at the bleaching time of 0.5h ...................... 192


bleached hairs for the second derivative at the bleaching time of 1h ......................... 193

List of Tables

13













Table 8.1: The reflectance data for 6% H202 bleached hair samples in the visible

spectrum ...................................................................................................................... 215

Table 8.2: The reflectance data for 9% commercial bleached hair samples in the

visible spectrum .......................................................................................................... 216

Table 8.3: The reflectance data for commercial persulphate bleached hair samples in

the visible spectrum..................................................................................................... 217

Table 8.4: Denaturation Temperatures and non-reversing Cp of virgin hair at various

heating rates for a typical DSC-curve ......................................................................... 218

Table 8.5: Denaturation Temperature and non-reversing Cp of Commercial Persulphate

Bleached hair (2h) at various heating rates for a typical DSC-curve .......................... 221

Abstract

14

Abstract

The University of Manchester

Daijiazi Zhang

Physical, Morphological and Chemical Structure & Property Relationships for α-

Keratins in Bleached Human Hair

02/07/2013

The surface and structural change of human hair fibre have been analysed to determine

the oxidation effects for bleached hairs. Three types of bleached hairs (6% H2O2 bleach,

9% H2O2 commercial bleach and commercial persulphate bleach (contains 9% H2O2))

as well as virgin hair were evaluated with the increasing treatment time using Scanning

Electron Microscopy (SEM), Reflective Spectrophotometry, Differential Scanning

Calorimetry (DSC) and Fourier Transform Infrared (FTIR) Spectroscopy. It is obvious

that longer treatment times result in the greater surface and structural damage.

However, commercial persulphate bleach causes less surface damage for the cuticle. 6%

H202 bleach has overall moderate damage effects on both cuticle and cortex over the

treatment time. 9% H2O2 commercial bleach indicates two different damage stages.

The first 1.5h bleached hairs show mild oxidation to the surface, whereas the damage

becomes heavy after 2h. This phenomenon results in that 9% H2O2 commercial bleach

has a more intensive oxidation damage in the cortex than the commercial persulphate

bleach. This is in line with DSC investigation which shows that the intermediate

filament of 9% H2O2 commercial bleach is heavily damaged after the extensive

oxidation time (≥2h). Although commercial persulphate bleach contains the stronger

oxidising agent, it has a less surface damage than 9% H2O2 commercial bleached hair

in FTIR-ATR measurement, and a similar oxidation effect on the matrix as 6% H202

bleached hair in FTIR transmission investigation. In addition, it has been verified by

colour measurements that bleached hairs have an overall lighter, yellowish and reddish

colour. Consequently, commercial persulphate bleached hair is much lighter and more

yellow than 9% H2O2 commercial bleached hair and 6% H202 bleached hair. DSC

investigations reveal that the three bleaches have a homogenous oxidation effect on IFs

and IFAPs. The deconvolution results using three Gaussian distributions confirm this

observation. The stronger bleach results in a homogenous structural damage on both

para- and ortho-cortex with increasing bleaching time. Commercial persulphate bleach

and 9% H2O2 commercial bleach have a progressive damage effect on the ortho- and

para- cortex than 6% H202 bleach. Kinetics analysis is conducted for the virgin and

bleached hairs by using various heating rates according to ASTM-E698. The activation

energies of 260 kJ/mol for the virgin hair and 295 kJ/mol for the commercial

persulphate bleached hair (2h) are determined from the slope of the regression line of

peak temperature, TD (as 1/TD) and heating rate, β (as lnβ) on the basis of the

Arrhenius-equation. The predominant structural damage for various heating rates only

occurs in the IF. It is shown that a linear increase in ∆HD occurs for lower heating rates,

while it is constant for higher heating rates. This can be ascribed to the hypothesis that

a lower heating rate favours a crystal transformation change (α-β transformation),

while a higher rate favours a crystalline-amorphous transformation. SEM examines the

morphological changes of hair samples after DSC. The cortex has been dissolved at the

lower heating rate. The commercial persulphate bleached hairs (2h) show an overall

shrunk cuticle surface and fewer and smaller hydrolysed protein granules, due to the

previous damage of the α-helix in the cortical cell.

Declaration

15

Declaration

No portion of the work referred to in the thesis has been submitted in support of an

application for another degree or qualification of this or any other university or other

institute of learning.

Copyright Statement

16

Copyright Statement

The author of this thesis owns certain copyright or related rights in it and has given

The University of Manchester certain rights to use such copyright, including for

administrative purposes.

Copies of this thesis, either in full or in extracts and whether in hard or electronic copy,

may be made only in accordance with the Copyright, Designs and Patents Act 1988

and regulations issued under it or, where appropriate, in accordance with licensing

agreements with the University. This page must form part of any such copies made.

The ownership of certain copyright, patents, designs, trade marks and other intellectual

property and any reproductions of copyright works in the thesis, for example graphs

and tables, which may be described in this thesis, may not be owned by the author and

may be owned by third parties. Such intellectual property and reproductions cannot

and must not be made available for use without the prior written permission of the

owner of the relevant intellectual property and/or reproduction.

Further information on the conditions under which disclosure, publication and

commercialisation of this thesis, the copyright and any intellectual property and/or

reproductions described in it may take place is available in the University IP Policy, in

any relevant thesis restriction declarations deposited in the University Library and in

the University’s policy on presentation of theses.

Acknowledgement

17

Acknowledgement

I would like to take this opportunity to express my gratitude to Prof. Franz J.

Wortmann and Dr. Gabriele Wortmann for their consistent encouragement and

spiritual support throughout all years of my PhD study. They not only mentored me

scientifically, but also stimulated me on “How to be a good researcher”.

I would like to give a big acknowledgment to Dr. Huw Owens for his kind assistance

in using the reflective spectrophotometer for the colour measurements, and to Mrs

Polly Crook for her technical advice for the thermal analysis by DSC. Moreover, I

would like to give special thanks to Mr. David Kenyon and Mr. Phil Cohen for their

help regarding the sample preparation in this work.

I am also grateful to the colleagues who worked in the same laboratory with me for

their scientific discussions and help.

I gratefully acknowledge the financial support during this project from Henkel AG

&Co KGaA, Hamburg, GER.

Last but not least, I would especially like to appreciate my family and friends. They

have supported me during the whole PhD project and have encouraged me to

overcome the difficulties in my life.

Abbreviations

18

Abbreviations

AFM Atomic Force Microscopy

ATR Attenuated Reflectance

Cer Ceramides

Ch Cholesterol

ChE Cholesterol Esters

ChS Cholesterol Sulphate

CIE Commission Internationale de L’Eclairage

CMC Cell Membrane Complex

DMA Dynamic Mechanical Analysis

DTGS Deuterated Triglycine Sulphate

DSC Differential Scanning Calorimetry

FFA Free Fatty Acids

FTIR Fourier Transform Infrared Spectroscopy

ΔHD Denaturation Enthalpy

ΔHD0

Denaturation Enthalpy of Virgin Hair

ΔHDR

Relative Denaturation Enthalpy

ICTA International Committee for Thermal Analysis

IF Intermediate Filament

IFAP Intermediate Filament-associated Protein

IR Infrared Spectroscopy

IRS Internal Reflection Spectroscopy

HGT Protein High Glycine/tyrosine-rich Protein

HS Protein High-sulphur Protein

Abbreviations

19

Hgt 0

Absolute Valley Depth of Cysteic Acid of Virgin Hair

Hgt R

Relative Value of Absolute Valley Depth of Cysteic Acid

KAP Keratin- associated Protein

LFM Lateral Force Microscopy

MDSC Modulated DSC

18- MEA 18-methyleicosanoic acid

MIR Mid-infrared

RH Relative Humidity

SAXS Small Angle X-ray Scattering

SEM Scanning Electron Microscopy

SNR Single-to-noise Ratio

SPD Spectral Power Distribution

TD Denaturation Temperature

TD0

Denaturation Temperature of Virgin Hair

TDR

Relative Denaturation Temperature

TEM Transmission Electron Microscopy

UHS Protein Ultra-high-sulphur Protein

XPS X- ray Photoelectron Spectroscopy

Chapter 1 Introduction

20


1.1 Morphology and Chemical Structure of Human Hair

Keratins are a group of specialised proteins containing a high content of cystine, which

has the capacity to cross-link the polypeptide chains by intermolecular disulphide

linkages. Structurally, keratins have been classified by two principal types of pattern,

namely the α-helical conformation for hard keratins of mammals, and the β-sheet

structure for those of birds and reptiles, respectively. Depending on how much cystine

they contain, the α-keratins are further subdivided into two groups, “hard” and “soft”.

The tougher material like horn, hoof and hair, which contains more than 3% sulphur, is

defined as hard keratin and flexible keratin such as stratum corneum in skin, which has

a lower sulphur content (less than 3%), is defined as soft keratin (Asquith, 1977).

Human hair is a keratin-containing appendage that grows from cavities in the skin,

called follicles. Hair follicles extend from the surface of the skin through the stratum

corneum and the epidermis into the dermis (Robbins, 2002). They are composed

mainly of fibrous proteins, which characteristically contain cystine residues, forming

intra- and intermolecular cross-links. The disulphide bond is one of the strongest bonds

in the natural world. The disulphide cross-links in hair keratin thus lead to high

mechanical strength and distinguish hair fibre from other fibrous proteins, such as silk

and collagen.

Human hair is a very complex fibre. Morphologically, a fully formed hair fibre (about

50-100 m in diameter) is primarily composed of three different structural units. From

the outside these are the cuticle, cortex and medulla, respectively. A schematic

representation of a human hair with its various layers of cellular structure is presented

in Figure 1.1.


21

Figure 1.1: Schematic diagram of a human hair with its various layers of cellular

structure (Feughelman, 1997)

1.1.1 Cuticle

The cuticle is the outermost protective layer, which is composed of flat, overlapping

scales that surround the central fibre core, or cortex. Each cuticle cell is approximately

0.5μm thick, with about 5 m exposed surface, and approximately 45-60 m long. In

human hair, the cuticle is generally 5-10 scales thick, so that the thickness of cuticle

layer is in the range of 2.5 to 5 m.

The cuticle is a barrier to the sorption of large molecules such as dyes, and protects the

cortex from physical damage (Feughelman, 1997). The cuticle has a scale-like

structure, in which the cuticle cells are attached at the root end and point towards the

tip of the hair fibre, like shingles on a roof. Each cuticle cell consists of a number of

sublamellar layers, namely, epicuticle, A-layer, exocuticle, and endocuticle. Individual

cells are separated by the cell membrane complex (Figure 1.2).


22

Figure 1.2: The fine structure of a human hair cuticle cell, which is separated from the

adjacent cell by the cell membrane complex (Swift, 2001b)

Table 1.1 describes the various layers of the cuticle cell and their respective cystine

level (Robbins, 2002). The epicuticle layer (about 3 nm thick) envelops the entire

cuticle cell and is a semipermeable, partly proteinaceous membrane, which allows the

diffusion of only small molecules into the cell (Ruetsch, 2001). It is covered with a

thin layer of 18-methyleicosanoic acid (18-MEA), which is covalently attached to

epicuticle proteins through thioester linkages involving cystine residues. This 18-MEA

layer makes up the outer β-layer of the cuticular cell membrane complex and is thus

responsible for low friction and provides a hydrophobic surface (Bhushan, 2010).

Underneath the outermost portion of the epicuticle is the A-layer of the cuticle cell

membrane, which has a very high cystine content (~30%). The A-layer is highly

crosslinked and acts as a biochemically stable layer, giving the cuticle surface

considerable mechanical toughness and chemical resilience.

The exocuticle (also known as the B-layer), which is immediately adjacent to the A-

layer, is also a cystine-rich component (~15%) forming about two-thirds of the scale

structure (Feughelman, 1997). On the inwardly facing side of each cuticle cell is a thin

layer of material which is referred to as the inner layer. Between the exocuticle and the

inner layer is the endocuticle which is deficient in cystine (~3%), and is mechanically


23

the weakest component of the cuticle structure. The endocuticle swells considerably

more in water than the exocuticle, due to their different cystine contents and thus the

corresponding cross-linking of the protein structure (Feughelman, 1997).

Table 1.1: The various layers of the cuticle, their cystine content and details about

their properties (Robbins, 2002)

Cuticle Layer Cystine Content (%) Details

Epicuticle ~12 18-MEA lipid layer attached to outer

epicuticle contributes to lubricity of the hair

A-layer ~30 Highly crosslinked

Exoxuticle ~15 Mechanically tough

Endocuticle ~3 Chemically resilient

Inner Layer

Cell Membrane

Complex

(CMC)

~2

Lamellar structure consists of inner β-layer,

δ-layer, and outer β-layer

1.1.2 Cell Membrane Complex (CMC)

The cell membrane complex (CMC), also known as intercellular matter or non-

keratinous region, is an adhesive material holding all the components of the hair

together, such as overlapping cuticle cells, cuticle and cortex cells, and neighbouring

cortex cells. It consists of a cell membrane and adhesive proteinaceous material. The

fraction of the CMC has been estimated to be about 3% (Bradbury, 1973). It is

believed that the cell membrane complex is the primary pathway for diffusion of

chemical agents (Robbins, 2002). The CMC itself has a lamellar structure, which is

made up of a central, polysaccharidic δ-layer, known as “intercellular cement”,

enclosed by two lipid-rich β-layers. The outer β-layer separates the cuticle cells from

each other, however, it is weakly bound to the δ-layer (Bhushan, 2010). The CMC is

primarily nonkeratinous protein, and is low in cystine content (< 2%), but high in polar

sulphur-containing amino acids (cysteine, thiocysteine). An important lipid component

of the CMC is 18-methyleicosanoic acid (18-MEA), which also appears on the outer

surface of each cuticle (upper layer) and is covalently connected to epicuticle protein


24

structures. Additionally, the CMC is also present in the cortex but with a variable

thickness of the δ-layer.

1.1.3 Cortex

The central hair fibre core, or cortex, which is surrounded by the protective layer of the

cuticle, makes up 90% of the hair mass. The cortex is largely responsible for the

physical and mechanical properties of the fibre and is composed of the cortical cells

and the cell membrane complex. The elongated, spindle-shaped cortical cells, are

generally 1 to 6 μm thick, approximately 100 μm long and run longitudinally in

parallel to the axis of the fibre.

The major structure within cortical cells are the closely packed macrofibrils, which

represent up to 60% of the cortex material by mass. The macrofibrils, approximately

0.1 to 0.4 m in diameter, consists of two main structures, microfibril and matrix,

which are distinguished by their structures and amino acid compositions. The

microfibril, or intermediate filament (IF), is a partly crystalline fibrous protein, which

is mainly composed of α-helical proteins with low content of cystine (6%). However

a gel-like matrix with a high cystine content (21%), acts as an embedding medium for

the intermediate filaments. The matrix varies in quantity and composition for different

keratins, whereas the microfibrils appear consistently from one keratin to another

(Feughelman, 1997). The matrix/intermediate filament ratio has been reported to be

unity in human hair (Robbins, 2002, Wolfram, 2003).

The matrix forms the largest structural portion of the cortex, is structurally less

organised and often classified as the amorphous region, although some evidence

suggests that it contains some structural organisation. Proteins of the matrix are

sometimes referred to as keratin-associated proteins (KAP) or intermediate filament-

associated proteins (IFAP). They are classified into three groups: high-sulphur (HS),

ultra-high-sulphur (UHS) and high glycine/tyrosine-rich (HGT) proteins.


25

The macrofibrils exhibit different variations in packing arrangement within the cortex.

Two or even three types of cortical cells are identified with different cystine (sulphur)

contents. These are the ortho-cortical cells, containing less matrix between the

intermediate filaments and a lower proportion of sulphur content (3%), para-cortical

cells with a higher concentration of disulphide linkages (5%), and mesocortical cells,

which show an intermediate cystine content between ortho- and para-cortical cells. The

cell groups may form two distinctive domains, e.g. in Merino wool called, respectively,

ortho-cortex (60-40%), which is more hydrophilic and heavily stained by some dyes,

and the para-cortex (40-10%), which is more readily stained by silver’s salts (Maclaren,

1981).

The composition of α-helical and coiled-coil segments in the intermediate filament is

summarised in Figure 1.3. Within the central rod domain, there are four main coiled-

coil segments 1A, 1B, 2A and 2B (Parry, 1985, Robbins, 2002). Each of these

segments is a left-handed two-stranded coiled-coil rope, which shows a hepta-peptide

sequence (heptade) repeat unit. The coiled coils are interrupted at three positions by

non-helical fragments (L1 links 1A to 1B, L2 links 2A to 2B and finally L12 links 1B

to 2A. The link L12 is about twice the size of L1 or L2) and are terminated by non-

helical N- and C- domains. A break in the heptade pattern, which involves some

distortion in the rope structure, is indicated in the centre of segment 2B. It was

suggested that the N- and C-terminal domains are likely to play a crucial role in

directing molecular and filament aggregation (Parry, 1985).

Figure 1.3: Schematic diagram of the rod domain of an intermediate filament dimer

(Parry, 1985)

Formation of the macrofibrils related to an aggregation process of intermediate

filaments is shown in Figure 1.4. The N- and C- terminal domains of each polypeptide

facilitate the assembly of coiled-coils (dimers) into protofilaments (tetramers), two of


26

which constitute a protofibril. Four protofibrils are packed to be a microfibril (7 to 8

protofilaments), which associates with other microfibrils to form a macrofibril.

Figure 1.4: Higher order -Keratin structure (Voet, 2008). (a) Two keratin

polypeptides form a dimeric coiled coil. (b) Two staggered coiled coils associated

head-to-tail to constitute protofilaments. (c) Two of the protofilaments assembled to

create a protofibril, four of which form a microfibril.

1.1.4 Medulla

Fine animal hairs (e.g., merino wool) have little or no medulla, but with increasing

fibre diameter, the medulla, which is the innermost layer of the hair fibre, can

sometimes be found. In human hair the medulla may be absent or, if present, may be

either continuous along the fibre axis, or discontinuous, and in some instances the

presence of a double medulla is observed.

Morphologically, the medulla is a 5 to 10 μm diameter channel of loosely packed and

largely empty cells forming an amorphous structure more centrally located in the

cortex. It has a porous structure formed by sponge-like keratin (Marhle, 1971, Nagase,

2002), which affects both colour and shine in white, brown and blond hair by

influencing the reflection of light.

The medulla has a high lipid and a cystine content. It is, however, rich in citrulline

(Kreplak, 2001a). A layer of CMC separates the medulla from the cortex (Swift,

1997a).


27

1.2 The Chemical Composition of Human Hair Fibres

It is important that human hair is an “integrated” system, with the chemical

components acting together. Depending on its moisture content, human hair consists

approximately 65–95% of keratin proteins, and the other constituents are water, lipids

(structural and free), pigments, and trace elements.

1.2.1 Proteins

Proteins are made up of long chains of various mixtures of some 20 to 50 amino acids

(Bhushan, 2010). Human hair is predominantly proteinaceous and has a group of

fibrous proteins known as α-keratin (Feughelman, 1997, Zviak, 1986). Like all

proteins, human hair proteins contain both cationic and anionic groups, and are

therefore amphoteric. The cationic character is due to the protonated side chains of

arginine, lysine, histidine, and the small portion of free amino groups at the ends of the

peptide chains. The characteristic of anionic groups is due to the side chains of aspartic,

glutamic acid and carboxyl end groups.

Figure 1.5 represents the condensation process of amino acids in proteins. The amino

acid residues are joined together by peptide bonds. After a number of these

condensation reactions it ultimately produces a high molecular weight protein, such as

keratin.

R2

HOOC

CHNH2

NH2HOOC

CH

R1

Figure 1.5: The condensation reaction of amino acids (Creighton, 1993)

Chemically, all ethnic hairs are found to have similar protein structure and composition,

(Menkart, 1984, Dekio, 1988a, Nappe, 1989, Dekio, 1990b). Although the quantities

and types of amino acids differ slightly between individuals, their ratios are very


28

similar. Table 1.2 gives an example of the amino acids present in hair and their relative

quantities.

Table 1.2: Typical amino acid composition of human hair. The analysis varies slightly

between individuals (Nishikawa, 1998)

It is obvious that the majority composition of amino acids in human hair is glutamic

acid, serine and cystine (shown as half-cystine), they make up 11.4%, 11.7% and 17.8%

of hair protein, respectively. Cystine is a dimeric amino acid formed by the oxidation

of two cysteine residues that covalently linked by two sulphur atoms, generating a very

strong bond, known as a disulphide linkage. Cystine residues provide the stability to

hair except during chemical processes, such as reduction, oxidisation, hydrolysis and

weathering. The differences in cystine content between various structures in human

hair result in the significant differences in their physical properties.

The primary chemical difference between bleached hair and virgin hair is a lower

cystine content, a higher cysteic acid content, and lower amounts of tyrosine and

methionine in the bleached hair (Robbins, 2002). This result is in good agreement with

Zahn’s original conclusion that the bleaching agents react with human hair protein

primarily at the disulphide bonds (Zahn, 1966).

It is interesting to observe that morphological components in hair can be classified as a

series of hierarchical two-phase composites (microfibril and intermicrofibrillar matrix;

Amino Acid % of Total Residues Amino Acid % of Total Residues

Lysine 2.7 Proline 8.4

Histidine 0.9 Glycine 6.4

Arginine 5.8 Alanine 4.6

Aspartate + Asparagine 4.9 Half-cystine 17.8

Threonine 6.8 Valine 5.8

Serine 11.7 Methionine 0.6

Glutamate + Glutamine 11.4 Isolucine 2.6

Leucine 5.8 Tyrosine 2.0

Phenylalanine 1.6

Total 99.8


29

macrofibril and intermacrofibrillar matrix; exocuticle and endocuticle; cortex and

cuticle). Different types of protein in the hair are found in these various phases (Swift,

1997b).

The cuticle of human hair contains a large percentage of cystine, cysteic acid, proline,

serine, threonine, isoleucine, methionine, leucine, tyrosine, phenylalanine, and arginine,

but only little tryptophan and histidine compared to the other layers of the hair shaft

(Bradbury, 1966). Also the cuticle includes a low but important amount of the fatty

acids, 18-methyleicosanoic acid (18-MEA), which is responsible for the water

repellent behaviour of the surface. And the protein portion of the cell membrane

complex (CMC) was found to be rich in the dicarboxylic amino acids, namely, aspartic

acid, and glutamic acid.

There is less cystine in the cortex than in the cuticle, but the cortex is richer in diacidic

amino acids, lysine, and histidine. In the cortex, the intermediate filaments and the

matrix are discriminated by their chemical compositions. The intermediate filaments

have a high content in leucine, glutamic acid, and those amino acids which are

generally present in α-helical proteins, but small amounts of cystine (6%), lysine, and

tyrosine (Fraser, 1972). Comparatively the matrix is rich in cystine (about 21%,

calculated from the sulphur content of γ-keratose of human hair), proline, and those

amino acids that are adverse to the helix formation such as in KAP proteins (Robbins,

2002).

The medulla has a poor solubility, and is thus difficult to isolate. Blackburn showed, in

a good agreement with Rogers’s results, that the medulla has a low cystine content and

relatively large amounts of acidic and basic amino acids (Blackburn, 1948, Roger,

1964). These observations suggest that the medulla will be more susceptible to

reactions with acids and alkalis and to ion exchange reactions such as reactions with

anionic and cationic surfactants, ionic dyes and metals. But the medulla will be less

sensitive to reaction with reducing agents (Robbins, 2002).

1.2.2 Water

Water is an important component of keratin fibres. Its content plays a critical role in

determining the physical and cosmetic properties of human hair (Robbins, 2002). The


30

moisture content in a keratin fibre depends on the relative humidity (RH) (Downes,

1961). A virgin hair can absorb more than 30% of its own weight of water, but it can

reach 45% when it is damaged. Hair length can increase by 2% and its diameter by 15%

to 20% through the absorption of water. It should be noted that the process of

absorption is very rapid, 75% of the maximum possible amount of water is absorbed

within 4 minutes.

Water is a highly polar molecule, it interacts with the hydrogen bonds and other polar

groups in the α-keratin chains. Two types of water are associated with the α-keratin

protein, absorbed or “bound” water and adsorbed or “free” water. At low humidities,

below 25%, water molecules are principally bonded to hydrophilic side chains

(guanidine, amino, carboxyl, phenolic, etc) and peptide bonds through hydrogen bonds

and Columbic interactions. With increasing humidity, additional water is absorbed,

water enters as “solution water”, resulting in a decrease in the energy of binding of

water already associated with the protein. At very high % RH (>80%), multi-molecular

sorption (water-on-water) occurs, and this refers to the “free” water interacting and

condensing onto the first “bound” layer (Robbins, 2002).

1.2.3 Lipids

Lipids from human hair mainly include cholesterol esters (ChE), free fatty acids (FFA),

cholesterol (Ch), ceramides (Cer) and cholesterol sulphate (ChS). Some of them (ChE

and FFA) result from sebum attributable to the sebaceous glands, whereas others (Ch,

Cer, ChS and additional FFA) are ascribed to the intrinsic constitutive lipids

biosynthesised in hair matrix cells (CMC). They are known as surface (external) lipids

and internal lipids, respectively. In addition, a part of the internal lipids is free lipid,

and another part is structural lipid of the cell membrane complex (Robbins, 2002).

Internal lipids are the principal constituent of cell membrane lipids (Hilterhaus-Bong,

1989, Körner, 1995), which contribute to the formation of a stable and strong CMC.

The CMC has a barrier function (Swartzendruber, 1989, Grubauer, 1989), and acts as a

protective layer, preventing external materials from penetrating the hair fibre.


31

Some researcher have confirmed that hair lipids have an influence on physicochemical

phenomena such as diffusion, cell cohesion and mechanical strength (Nishimura, 1989,

Sideris, 1990, Braida, 1994, Philippe, 1995), despite their low content (1–9% dry

weight) compared to proteins (>90%).

1.2.4 Pigments

Hair pigments, which give the colour to hair, are mainly concentrated in the cortex as

melanins. Melanin is produced by a group of specialised cells called melanocytes.

These cells exist near the hair bulb and are present in the form of discrete granules,

referred to as pigment granules. Pheomelanin and eumelanin are the two types of basic

pigments. Their chemical structures are presented in Figure 1.6.

Figure 1.6: Chemical structures of eumelanins and pheomelanins (Colbert, 2004).

Curly red lines indicate sites of attachment to the extended polymer or proteins.

Pheomelanin is the most common melanin in the Caucasian hair, which is responsible

for the lighter colours such as red and fair. It is soluble in aqueous alkali,

trifluoroacetic acid, formic acid, and several highly polar organic solvents. Eumelanin,

which produces the dark shades such as brown and black, contains nitrogen (6-9%) but

insignificant amounts of sulphur (0-1%) and is insoluble in solvents and chemically

resistant to all but powerful oxidizing agents, such as hydrogen peroxide. The ratio

between two pigments determines the hair colour (Robbins, 2002). Vincensi et al.

(1998) studied red hair, observing that the amounts of pheomelanin and eumelanin

vary with sex, age and colour shade. However, the absence of either type of melanin

results in white hair.


32

1.2.5 Chemical Elements

In terms of the chemical elements, hair is composed of 50.7% carbon, 20.9% oxygen,

17.1% nitrogen, 6.4% hydrogen and 5.0% sulphur. Trace elements contained in the

hair are magnesium, arsenic, iron, chromium and other metals and minerals. Most of

them are incorporated in the hair from extraneous sources and probably integrated into

the fibre structure by salt linkages or as coordination complexes with side chains of

pigments and/or proteins.

1.3 Bonding Mechanisms in Keratin

In the α-keratin structure, stability is determined by a variety of bonding mechanisms,

which is shown in Figure 1.7. Except the strong covalent disulphide bonds, other

weaker interactions such as coulombic interactions between side chain groups,

hydrogen bonds between neighbouring groups, van der Waal’s interactions,

isodipeptide bridges between segments of two hypothetical peptide chains and, in the

presence of water, hydrophobic bonds also contribute to its stability (Zviak, 1986,

Feughelman, 1997, Johnson, 1997).

The covalent disulphide (-S-S) linkages are formed by an oxidation reaction between

adjacent thiol (-S-H) groups of opposing cysteine molecules in the polypeptide chain,

thereby resulting in the formation of cystine (Johnson, 1997, Williams, 1994), which

contributes significantly to the physical and chemical properties of hair keratin

(Feughelman, 1997).

Coulombic interactions, also refer as salt links, are electrostatic forces, acting between

ionised acidic and basic side chain residues, i.e. the negatively charged carboxylic acid

groups (─COO─) and positively charged amino groups (─NH3

+) (Zviak, 1986,

Johnson, 1997, Feughelman, 1997). These charged groups exist under neutral pH

conditions. At high pHs with an excess of OH─

present ─NH3+ converts to ─NH2

while at low pHs with an excess of hydrogen ions present ─COO─ forms ─COOH.

The coulombic interactions thus disappear at both above conditions.


33

Figure 1.7: Structural formula of a fictive peptide chain to illustrate five important

interactions between amino acid side-chains in keratins (Zahn, 2003).

There are two types of hydrogen bonding which exist in the α-keratin structure. One

type (─O…H─O─) corresponds to the bonding between water molecules and hydroxyl

groups and another type (─O…H─N─) is present between the amide ─NH and the

carbonyl group as well as the amide C=O of side chains (see Figure 1.8). Hydrogen

bonds play a central role for the stabilisation of the α-helical structure in keratin.

Figure 1.8: Typical hydrogen bonds involving the atoms of oxygen and nitrogen

(Feughelman, 1997)

Hydrogen Bond

Hydrophobic Bond

Disulphide Bond

Coulombic Interaction

Isodipeptide Bond


34

Van der Waal’s forces (or van der Waal’s interaction), are involved in the cohesive

bonding of the chains and side chains of α-keratin fibres, which provide molecular

stability. Finally, hydrophobic bonds (only in the presence of water) have the

specialised capacity of binding the single α-helices into double α-helical ropes, which

ultimately form intermediate filaments (Feughelman, 1997). Both van der Waal’s

forces and hydrophobic bonds have a non-directional action. Since they are long-range

weak forces, these bonds are susceptible to breakdown and reformation.

An isodipeptide bridge is formed by the reaction between glutamic acid and lysine. A

second covalent bridge, though relatively rare, is caused by the isodipeptide Nε-(γ-

glutamyl) lysine residue which provides an additional stabilising effect in the resistant

cell membranes and the cuticle (Zahn, 2005).

1.4 The Keratin-Water System

Feughelman (1959) suggested a 2-phase model for the keratin fibre, in which water-

impermeable microfibrils (phase C) are embedded in a water-penetrable matrix (phase

M), as shown in Figure 1.9. This two-phase model plays an important role in

explaining the chemical and physical properties of α-keratin fibres and their changes

with moisture content.

Figure 1.9: Cylindrical two-phase model of an -keratin fiber consisting of rigid

water-impenetrable phase C set parallel to the fiber axis, embedded in the matrix phase

M, which is water-penetrable (Feughelman, 1959).


35

Due to the two-phase model, a dried hair, which is absorbing water from the liquid

state, must continuously adjust its swollen outer annular layer to adapt to the swelling

which happens inside the cylindrical fibre. The water penetrable phase M swells in the

presence of water, whereas the water impenetrable phase C stays unchanged.

Phase C has been identified as the crystalline α-helical structure present in the

microfibrils or filaments. Since phase C is arranged parallel to the fibre axis for the

fibre, swelling in water mainly occurs in the diametral direction.

1.5 Chemical Treatments

Common hair chemical treatments include reducing hair, such as permanent waving,

and oxidizing hair, such as bleaching and dying. A permanent wave treatment reduces

the cystine disulphide bond, and then reforms the bonds with the oxidising agent.

Bleaching and dying treatments oxidise cystine and other amino acids with alkaline

hydrogen peroxide. Hair damage caused by these chemical treatments, altering the

chemical and physical properties of the α-keratin fibre, can be discriminated from

untreated fibre. Currently, bleaching damage effects on hair samples in terms of

physical and chemical properties are of particular interest.

1.5.1 Bleaching Principle

The objective of bleaching is to oxidise the hair pigments in the cortex and thereby to

lighten the natural hair shade by an irreversible chemical reaction. Bleaching relies

upon oxidative destruction of the highly conjugated system of the indole residues of

the melanins. The melanin still exists in the bleached hair, but appears colourless. The

bleaching agent reacts more readily with the dark eumelanin pigment than with the

pheomelanin, so some gold or red residual colour may remain. Additionally, bleached

hair displays a pale yellow tint due to the natural yellow colour of keratin (Robbins,

2002).


36

Bleaching is applied through a mixture of several components, where the primary

chemical is hydrogen peroxide. Hydrogen peroxide, as the most common oxidising

agent, is adopted because it oxidises the melanin faster than proteins so colour

lightening can be achieved faster to prevent further protein damage (Robbins, 2002).

Generally, the higher the hydrogen peroxide concentration, the greater the changes in

hair colour. A bleaching formula usually contains 6 - 12% of hydrogen peroxide

combined with a booster such as ammonium and potassium persulphates. Ammonia is

employed to open the outer layer of the hair cuticle, which allows the peroxide to

oxidise the melanin. The pH is around 10 and stabilisers (e.g. sequestrants) are often

used to reduce the rate of decomposition of the peroxide to provide satisfactory shelf

life. During bleaching some keratin dissolves, however the weight losses are very

small (Wolfram, 1970b).

Wolfram et al. (1975) confirmed that the melanin pigments encapsulated within the

granules are not accessible to oxidising agents, unless the granules are degraded or

solubilised. The bleaching process follows two steps, first, a fast dissolution step

occurs in which the pigment granules are dispersed and dissolved by hydrogen

peroxide. Hydrogen peroxide is not as strong an oxidising agent as permanganate or

peracetic acid, but it is actually more effective for dissolving the granules (Robbins,

2002). It is worth pointing out that the disintegration of the pigment granules and their

solubilisation are not likely to affect the colour of hair significantly (Wolfram, 1970b).

After the granules are dissolved, this is then followed by a much slower decolouration

step, which determines the overall bleaching rate. The pigment granules are distributed

within the cortex. As a consequence of the decolouration of the melanin, an oxidation

effect also takes place in the keratin, and this is often referred to as “oxidative or

bleaching” damage (Wolfram, 1970b). After the progressive treatment time, the

melanin granules were dissolved by the peroxide, leaving most of the granular cavities

empty. This makes hairs more porous which will hence absorb larger amounts of water

(Johnson, 1997). Besides, although persulphate is not as strong an oxidant as hydrogen

peroxide, persulphate mixed with peroxide is more effective than peroxide alone

(Robbins, 2002). Thus, persulphate and peroxide complement each other in terms of

their ability to bleach melanin pigment in human hair.


37

As described previously (Section 1.2.1), hair is primarily a proteinaceous fibre.

Strongly oxidising alkaline solutions thus not only act on the melanin itself, but also

attack the accessible reaction sites of the protein. This results in weakening of the cell

membrane complex, oxidation of the cystine residues of the matrix in the cortex and of

other hair components, which are rich in cystine, such as A-layer and exocuticle. As a

result, the bleached fibre becomes more brittle, drier, rougher and weaker, and is more

susceptible to breakage. In addition, bleaching leads to a lower cross-link density and

thus increases any hydrophilic property, due to anionic site formation e.g. cysteic acid

residue (Jachowicz, 1987, Johnson, 1997).

1.5.1.1 The Disulphide (S-S) Cleavage Mechanism

The bleaching reaction between keratin and H202 is defined mainly by the cystine

residues. The decrease in cystine is almost quantitatively matched by a corresponding

increase in cysteic acid. Figure 1.10 presents the oxidation cleavage of the disulphide

bond during the chemical bleaching. Complete oxidation of cystine residues yields

cysteic acid residues or sulphonic acid residues (-SO3H), but other oxidative

intermediates are also formed under controlled conditions of oxidation, such as the

cystine monoxide (-SO-S-) and cystine dioxide (-SO2-S-).

S

S

R

R

S

S

R

R

O

S

S

R

R

O

O

Cystine Residues Cystine Monoxide Cystine Dioxide

S

O

O

OHR

S

O

O

OHR'

Cysteic Acid Residues/Sulphonic Acid Residues

Figure 1.10: Scheme of the S-S cleavage mechanism for the bleaching process

(Robbins, 1971)

+ H2O2 + H2O2 + H2O2

or R/R’: side-chain of different structure


38

1.6 Objectives

The aim of this PhD study is to examine keratin surface and structural change caused

by various bleaching treatments (oxidation) and to interpret the interrelations between

the properties by numerous methods. In this work, untreated and three types of

bleached human hairs (6% H202 bleach, 9% H202 commercial bleach and commercial

persulphate bleach (contains 9% H2O2)) were characterised by a multidisciplinary

approach using Differential Scanning Calorimetry (DSC), Fourier Transform Infra-Red

Spectroscopy (FTIR) with Attenuated Total Reflectance technique (ATR) and

Transmission, Scanning Electron Microscopy (SEM) and Reflective

spectrophotometry.

Reflective spectrophotometry is employed to determine the colour changes of bleached

hair.

Thermal analysis by DSC will be used to measure the denaturation and decomposition

behaviour of both untreated hair and treated hair over a range of temperatures

compared to a reference. A kinetic study of the denaturation process is also performed.

Attenuated reflectance (ATR) and transmission modes of FTIR spectroscopy are

introduced to characterise and quantify oxidation of treated hair of cysteic acid with

respect to the formation.

The morphological changes of the cuticle surface topography as a consequence of

damage processes are assessed by means of SEM.

The data obtained from each method will be compared to illustrate how different

oxidation treatments affected the hair samples. There are various treatment factors

which affect the oxidation results, such as temperature, pH, peroxide volume, humidity

etc, but the effect of bleaching time was the most obvious factor when compared with

other bleaching parameters (Yuen, 2009), thus the investigation of oxidative duration

is of special interest.


39

1.7 Reference

ASQUITH, R. S. 1977. Chemistry of Natural Protein Fibers, New York, Plenum Press.

BHUSHAN, B. 2010. Biophysics of Human Hair: Structural, Nanomechanical, and

Nanotribological Studies, Springer.

BLACKBURN, S. 1948. The composition and reactivity of medullated keratin.

Biochemistry Journal, 43, 114-117.

BRADBURY, J. H. 1973. The Structure and Chemistry of Keratin Fibers. Advances in

Protein Chemistry, 27, 111-211.

BRADBURY, J. H., CHAPMAN, G.V., HAMBLY, A.N., KING, N.L.R., 1966.

Separation of Chemically Unmodified Histological Components of Keratin

Fibres and Analyses of Cuticles. Nature, 210, 1333-1334.

BRAIDA, D., DUBIEF, C., LANG, G., HALLEGOT, P., 1994. Ceramide: a new

approach to hair protection and conditioning. Cosmetics and toiletries, 109, 49-

57.

COLBERT, A., SCHOLAR, M., 2004. Towards Probing Skin Cancer using

Endogenous Melanin Fluorescence. The Penn State Manair Journal 11, 8-15.

CREIGHTON, T. E. 1993. Proteins: Structures and Molecular Properties, New York,

W.H.Freeman and Company.

DEKIO, S., JIDOI, J., 1988. Hair low-sulfur protein composition does not differ

electrophoretically among different races. The Journal of Dermatology, 15,

393-396.

DEKIO, S., JIDOI, J., 1990. Amount of fibrous and matrix substances from hair of

different races. The Journal of Dermatology, 19, 62-64.

DOWNES, J. G. 1961. The Attainment of a Defined State of Dryness in Accurate

Determinations of the Regain of Wool Samples. Textile Reseach Journal, 31,

66.

FEUGHELMAN, M. 1959. A Two-Phase Structure for Keratin Fibers. Textile Reseach

Journal, 29, 223.

FEUGHELMAN, M. 1997. Mechanical Proeprties and Structure of Alpha-Keratin

Fibres: Wool, human hair and related fibres, University of New South Wales

Press.

FRASER, R. D. B., MACRAE, T.P., ROGERS, G.E., 1972. Keratins : their

composition, structure and biosynthesis, USA, Charles C Thomas.


40

GRUBAUER, G., FEINGOLD, K.R., HARRIS, R.M., ELIAS, P.M., 1989. Lipid

content and lipid type as determinants of the epidermal permeability barrier.

Journal of Lipid Research, 30, 89-96.

HALL, K., WOLFRAM, L.J., 1975. Isolation and identification of the hair protein

component of hair melanin. Journal Cosmetic Chemists, 26, 247.

HILTERHAUS-BONG, S., ZAHN, H., 1989. Contributions to the chemistry of human

hair. II. Lipid chemical aspects of permanently waved hair. International

Journal of Cosmetic Science, 11, 167-174.

JACHOWICZ, J. 1987. Hair damage and attempts to its repair. Journal of the Society

of Cosmetic Chemists, 38, 263-286.

JOHNSON, D. H. 1997. Hair and Hair Care, New York, Marcel Dekker, Inc,.

KÖRNER, A., PETROVIC, S., HOCKER, H., 1995. Cell Membrane Lipids of Wool

and Human Hair Form Liposomes. Textile Research Journal 65, 56-58.

KREPLAK, L., BRIKI, F., DUVAULT, Y., DOUCET, J., MERIGOUX, C., LEROY,

F., LÉVÊQUE, J. L., MILLER, L., CARR,G. L., WILLIAMS, G. P.,

DUMAS, P., 2001. Profiling lipids across Caucasian and Afro-American hair

transverse cuts, using synchrotron infrared microspectrometry. International

Journal of Cosmetic Science, 23, 369-374.

MACLAREN, J. A., MILLIGAN, B., 1981. Wool Science: The chemical reactivity of

the wool fibre, Merrickville, Science Press.

MARHLE, G., ORFANOS, G.E., 1971. The spongious keratin and the medulla of

human hair. Archives of Dermatological Research, 241, 305-316.

MENKART, J., WOLFRAM, L.J., MAO, I., 1984. Caucasian hair, Negro hair, and

wool: similarities and differences. Journal Society Cosmetic Chemists, 35, 21-

43.

NAGASE, S., SHIBUICHI, S., ANDO, K., KARIYA, E., SATOH, N., 2002. Influence

of internal structures of hair fiber on hair appearance. I. Light scattering from

the porous structure of the medulla of human hair. Journal of Cosmetic Science,

53, 89-100.

NAPPE, C., KERMICI, M., 1989. Electrophoretic analysis of alkylated proteins of

human hair from various ethnic groups. Journal Society Cosmetic Chemists, 40,

91-99.

NISHIKAWA, N., TANIZAWA, Y., TANAKA, S., HORIGUCHI, Y., MATSUNO,

H., ASAKURA,T., 1998. PH Dependence of the Coiled-coil Structure of

Keratin Intermediate Filament in Human Hair by 13C NMR Spectroscopy and

the Mechanism of Its Disruption. Polymer Journal 30, 125-132.


41

NISHIMURA, K., NISHINO, M., INAOKA, Y., KITADA, Y., FUKUSHIMA, M.,

1989. Interrelationship between the hair lipids and the hair moisture. Nippon

Koshohin Kagakkaishi, 13, 134-139.

PARRY, D. A. D., FRASER, R.D.B., 1985. Intermediate filament structure: 1.

Analysis of IF protein sequence data. International Journal of Biological

Macromolecules, 7, 203-213.

PHILIPPE, M., GARSON, J.C., GILARD, P., HOCQUAUX, M., HUSSLER, G.,

LEROY, F., MAHIEU, C., SEMERIA, D., VANLERBERGHE, G., 1995.

Synthesis of 2-N-oleoylamino-octadecane-l,3- diol: a new ceramide highly

effective for the treatment of skin and hair. International Journal of Cosmetic

Science 17, 133-146.

ROBBINS, C. R. 1971. Chemical Aspects of Bleaching Human Hair. Journal of the

Society of Cosmetic Chemists, , 22, 339-348.

ROBBINS, C. R. 2002. Chemical and Physical Behavior of Human Hair, New York,,

Springer-Verlag.

ROGER, G. E. 1964. Structural and Biochemical features of the Hair Follicle. In:

MONTAGNA, W., LOBITZ, W. C., (ed.) The epidermis. New York:

Academic Press.

RUETSCH, S. B., KAMATH, Y., WEIGMANN, H.D., 2001. Photodegradation of

Human Hair: A Microscopy Study. In: GIACOMONI, P. U. (ed.) Sun

Protection in Man. Amsterdam: Elsevier Science.

SIDERIS, V., HOLT, L. A., LEAVER, I. H., 1990. A microscopical study of the

pathway for diffusion of rhodamine B and octadecylrhodamine B into wool

fibres. Journal of the Society of Dyers and Colourists,, 106, 131-135.

SWARTZENDRUBER, D. C., WERTZ, P. W., KITKO, D. J., MADISON, K. C.,

DOWNING, D. T., 1989. Molecular models of the Intercellular Lipid Lamellae

in Mammalian Stratum Corneum. Journal of Investigative Dermatology, 92,

251–257.

SWIFT, J. A. 1997a. Fundamentals of Human Hair Science, Weymouth Dorset,

Micelle Press.

SWIFT, J. A. 1997b. Morphology and histochemistry of human hair. In: JOLLES, P.,

ZAHN, H., HOCKER, H., (ed.) Formation and Structure of Human Hair.

Berlin: Birkhauser Verlag.

SWIFT, J. A., SMITH, J. R., 2001. Microscopical investigations on the epicuticle of

mammalian keratin fibres. Journal of Microscopy, 204, 203-211.

VINCENSI, M. R., D'LSCHIA, M., NAPOLITANO, A., PROCACCINI, E. M.,

RICCIO, G., MONFRECOLA, G., SANTOIANNI, P., PROTA, G., 1998.

Phaeomelanin versus eumelanin as a chemical indicator of ultraviolet


42

sensitivity in fair-skinned subjects at high risk for melanoma: a pilot study.

Melanoma Research,, 8.

VOET, D., VOET, J.G., PRATT, C. W., 2008. Fundamentals of Biochemistry: Life at

the molecular level, New York, John Wiley & Sons.

WILLIAMS, A. C., EDWARDS, H. G. M., BARRY, B. W., 1994. Raman spectra of

human keratotic biopolymers: Skin, callus, hair and nail. Journal of Raman

Spectroscopy, 25, 95-98.

WOLFRAM, L. J. 2003. Human Hair: A unique physicochemical composite. Joural

American Academy of Dermatology, 48, 106-114.

WOLFRAM, L. J., HALL, K., HUI, I., 1970. The Mechanism of Hair Bleaching. J.

Soc. Cosmet. Chem, 21, 875-900.

YUEN, C. W. M., KAN, C.W., LAU, K.W., CHOW, Y. L., 2009. Effect of Different

Human Hair Bleaching Conditions on the Hair coloration with Hair Boosting

Shampoo as Colorant. Fibers and Polymers, 10, 709-715.

ZAHN, H., WORTMANN, F. J., WORTMANN, G., SCHAEFER, K., HOFFMANN,

R., FINCH, R., 2003. Wool. Ullmann’s Encyclopedia of Industrial Chemistry.

Weinheim, Germany: Wiley-VCH Verlag.

ZAHN, H., WORTMANN, F. J., HÖCKER, H., 2005. Considerations on the

occurrence of loricrin and involucrin in the cell envelope of wool cuticle cells.

International journal of sheep and wool science, 53, 2.

ZAHN, H. J. 1966. Chemical Vorgiinge in the bleaching of wool and human hair with

hydrogen peroxide and peroxysiiuren. Society of Cosmetic Chemists, 17, 687-

701.

ZVIAK, C. 1986. The Science of Hair Care, New York, Marcel Dekker.

Chapter 2 Sample Preparation

43


2.1 Sample Material

Two batches of virgin Caucasian commercial hairs were used for the studies (Kerling,

Germany) to perform the oxidation reaction, namely hair batch I (13cm long) and hair

batch II (18cm long). Three treatments of the hair samples were studied and are

referred as: 6% H2O2 bleached hair, 9% H2O2 commercial bleached hair, and

commercial persulphate (contains 9% H2O2) bleached hair. Particularly, hair batch I

was used for the oxidative treatment of 6% H2O2 and 9% H2O2 commercial bleach, and

hair batch II was introduced for the use of commercial persulphate bleach (contains 9%

H2O2). Hairs are oxidised for bleaching times of 1/2h, 1h, 1½h, 2h, 2½h, 3h, 3½h and

4h, respectively, with fresh bleaching solution every half an hour.

2.2 Bleaching Products & Procedures

2.2.1 Bleaching Products

2.2.1.1 6% H2O2 Bleaching

6% laboratory-grade H2O2 was prepared; pH was adjusted with NH4OH to around 10.2

in this experiment. All bleaching treatments were done at room temperature with a

liquid: fibre ratio of 400:1 (hair tress ~ 250mg, bleaching solution ~ 100ml)

2.2.1.2 9% H2O2 Commercial Bleaching

9% H2O2 commercial bleaching developer was prepared (Henkel, Germany); pH was

adjusted with NH4OH to around 10.2 in this experiment. The commercial developer

recipe is described as below:


44

Developer ( IGORA Royal 20 vol 9% H202 bleaching lotion, Schwarzkopf)

Aqua (Water), Hydrogen Peroxide (9%), Cetearyl Alcohol, Propylene Glycol,

Ceteareth-20, Steartrimonium Chloride, Paraffinum Liquidum (Mineral oil), Etidronic

Acid, 2,6-Dicarboxypyridine, Disodium Pyrophosphate, Pottassium Hydroxide,

Parfum (Fragrance), Sodium Benzoate

2.2.1.3 Commercial Persulphate Bleaching

The bleaching powder and 9% H2O2 commercial bleaching developer (Henkel,

Hamburg, Germany) were mixed with a ratio of 1:2. The bleach powder recipe is

described as below:

Dust-free bleach powder for controlled lightening (IGORA Vario Blond Plus

bleaching powder, Schwarzkopf)

Potassium Persulphate, Magnesium Carbonate Hydroxide, Sodium Metasilicate,

Paraffinum Liquidum, Ammonium Persulphate, Cellulose Gum, Silica, Acrylates

Copolymer, Triticum Vulgare Starch, Disodium EDTA, Calcium Stearate, Hydrolysed

Keratin, Sodium Starch Octenylsuccinate, Sodium Hexametaphosphate, Parfum, CI

77007

2.2.2 Bleaching Procedure

2.2.2.1 6% H2O2 Bleaching

The hair samples were bleached in the prepared solution. Once the oxidative treatment

was accomplished, bleached samples were rinsed immediately under running tap water

for 3-5 minutes, subsequently 500ml tap water were prepared and the rinsing water

was changed every 30 mins. The water pH was recorded before and after this change.

This process aimed to remove unfixed bleach from the hair. When the value of pH was

stable, the hair samples were dried under ambient room conditions overnight before

further investigations.


45

2.2.2.2 9% H2O2 Commercial Bleaching

The freshly prepared bleaching solution was applied to the dry hair sample evenly

from root to tip using a brush. Each hair sample was covered by aluminium foil to

avoid evaporation and the chemical reaction with air. The bleaching solution was

changed every 30 mins to achieve optimum bleaching effect.

Samples were rinsed with warm water for 5 mins between each change of the

bleaching solution until the required treatment time had been achieved. After the whole

procedure, the hair tresses were rinsed with warm water to remove all the remaining

bleaching mixture for 5 mins, and then were dried under ambient room conditions

overnight before further investigations.

In addition, hair samples, which were bleached for 3½h and 4h, were rinsed with water

overnight.

2.2.2.3 Commercial Persulphate Bleaching

The required amount of bleaching powder and bleaching developer were put into a

bowl and mixed together before use. The bleaching mixture was applied to hair evenly

from root to tip using a brush. Each hair sample was covered by aluminium foil to

avoid evaporation and the chemical reaction with air. The bleaching paste was changed

every 30 mins to achieve optimum bleaching effects. The hair tresses were turned over

every 6 mins to assist the even application of the cream.

Samples were rinsed with warm water for 5 mins between each change of bleaching

cream until the required treatment time had been achieved. After the whole procedure,

the hair tresses were rinsed with warm tap water to remove all the remaining bleaching

mixture for 5 mins and were dried at ambient room temperature overnight before

further investigations.

In addition, hair samples, which were bleached for 3½h and 4h, were rinsed with water

overnight.

Chapter 3 SEM Investigation of the Surface Properties of Human Hair

46

Chapter 3 SEM Investigation of the Surface

Properties of Human Hair

3.1 Introduction

Although the cuticle accounts for only 10% by weight of the hair fibre, it plays a

protective role for the cortex against mechanical, chemical and thermal damage (Swift,

1999) and is responsible for the surface properties of the fibre. Thus the effect of

cosmetic treatments on the cuticle has been of great interest for the cosmetics industry

(Robbins, 1994).

3.1.1 Evaluation Method for Surface Properties

Numerous techniques have been used to study the properties of the hair cuticle,

including scanning electron microscopy (SEM) (Poletti, 2003), transmission electron

microscopy (TEM) (Swift, 2001a), X-ray photoelectron spectroscopy (XPS) (Dalton,

2000), atomic force microscopy (AFM) (Smith, 1998, Poletti, 2003, Gurden, 2004),

confocal microscopy (Corcuff, 1993), microdiffraction (Kreplak, 2001b), secondary

ion mass spectrometry (Gillen, 1999), goniometry (Feughelman, 2001) and lateral

force microscopy (LFM) (McMullen, 2001).

3.1.1.1 Scanning Electron Microscopy (SEM)

SEM is the standard method for examining the surface topography of human hair.

SEM images, using the electrons reflected from a specimen, provide topographical

(surface features of an object), morphological (shape and size of the particles making

up the object) and compositional (the elements and compounds that the object is

composed of and the relative amounts of them) information regarding the hair sample.


47

Because the SEM generally utilises vacuum conditions and uses electrons to form an

image, sample preparation is necessarily required for SEM measurements. Since hair

fibres are a non-metal material, it must be covered with a very thin layer of a

conductive material to prevent charging by the electrons. This coating process is called

“sputter coating”. Both metallisation and measurement needs to be operated under

vacuum, but surface metallisation and vacuum exposure could potentially cause

modifications to the surface details (Poletti, 2003).

3.1.1.1.1 Advantages and Disadvantages of SEM

The SEM was chosen for this study because its micrographs contain much more

topographical information than other microscopic techniques (Dibianca, 1973). The

produced images reveal surface structural details over a wide range of magnifications.

Even at a relatively low magnification, the SEM still has advantages for hair

characterisation compared to light microscopy, such as a clear view of cuticle uplift or

surface breakage (Dibianca, 1973).

Most SEMs are comparatively easy to operate. In addition, its relatively short data

acquisition time (less than 5 minutes/image), highly portable digital data format, much

enhanced resolution, a large depth of focus and the greater magnification, result in the

SEM having substantial advantage over other microscopic techniques. SEM is,

however, not able to reproduce colour and can only provide limited quantitative data

regarding the surface.

Light microscopy has an advantage over electron microscopy that it can be used for

observing internal structures, i.e. medulla and pigmentation, which are also very

individual features for hair fibres (Kadikis, 1987, Palenik, 1983, Sich, 1990). As a

result, Sich (1990) concluded that electron microscopy and light microscopy could be

employed in a complementary manner for the study of topographical features and

internal structures of hair fibres, respectively.


48

3.1.2 The SEM Investigation of Human Hair

The surface characteristics of the cuticle cells, such as scale shape, height, and

distribution, are considered to be unique to each hair fibre, thus, SEM is employed to

identify, distinguish and classify the different keratin fibres by measuring the step

height (the height of one cuticle relative to the adjacent one) of the cuticle scale edge.

(Kadikis, 1987, Langley, 1981, Wortmann, 1986).

However, one of the main problems of the analysis of human hair is the difficulty of

representative sampling, because the variability of the surface architecture, distribution

and appearance of the scales for hairs from one head are already indefinite, due to the

natural and cosmetic history (Dibianca, 1973). Additionally, variation is caused by

differences in the region of the head, from which they originate, and differences along

the length of the fibre from root to tip, as a result of the natural weathering process

(e.g. exposure to sunlight) (Brown, 1974) and the personal grooming habits (e.g.

brushing), to which hairs are subjected (Gurden, 2004). Consequently, the use of

imaging to determine the change of the surface on the hair should consider two factors:

(a) compare hair samples from the same regions and (b) measure as many images as

necessary to achieve a representative description (Gurden, 2004).

3.2 Objectives

The investigation of the surface topography of both untreated and chemically treated

human hairs at a microscopic level has provided a basis for the understanding of

damaged keratin fibre on a structural level. The prime objective of the SEM

investigation is to explore a method for establishing the extent of damage from

oxidative treatments. For the surface property investigation, the feasibility of the “Knot

Test” method is investigated to assess numerically the effects of the various bleaching

treatments on the conditions of the hairs. The morphological changes of hair fibres

were determined by the “Knot Test” methodology as a function of the severity of the

treatment of the hair.

For this study, three types of hair fibre were analysed:

(1) Untreated fibres with little damage to the cuticle


49

(2) Mildly bleached fibres, which show moderate chipping and jaggedness of the

cuticle edges

(3) Heavily bleached fibres, which display the largest extent of damage to the

cuticle

3.3 Experimental

Swift and Smith (2000) verified that different features of the hair fibre surface

gradually change along its length from root to the tip of the fibre. Although hair length

and diameter varied in all hair samples, the characteristic morphology showed little

variation for any individuals. Virgin and three types of bleached hairs were examined

in this study in order to investigate the differences between complete and weakened

cuticle cells. The untreated hairs are considered as standard and are defined as having

an intact cuticle and absence of chemical damage. The bleaching conditions are

described in the Chapter 2.

Most human hairs are assumed to have a similar damage stage in the middle part of the

hair length, thus the SEM experiments were conducted on the middle parts of the hair

samples. Ten fibres per tress were randomly selected and knotted, making sure to

achieve tight knots without exerting excessive force. Since there are three hair tresses

in each treatment time, 30 hair fibres per treatment time were knotted. Hair fibres then

were securely mounted on metal stubs using double sided adhesive tape. 30 virgin

hairs also were chosen and knotted as a reference.

Due to the instrumental limitations and the nonconductive nature of hair, the work is

undertaken after applying a metallic conductive gold coating. The samples were

vacuum sputter coated with a thin layer of gold (30nm) and kept in high vacuum (-10-4

Torr) for 1 minute. This process was repeated 3 times using the Sputter Coater 91000

(Edwards, UK). SEM images were obtained with the scanning electron microscopy

SM 300 (Topcon, UK) at 5 kV accelerating voltage and 8x spot size. 300x as a

standard magnification was chosen for all the knot test investigations.


50

3.4 Results and Discussions

3.4.1 Morphological Structures of Bleached Hairs

Oxidation causes structural damage to hair fibres. The SEM technique was used to

examine the surface of hair fibres after they had been subjected to various specified

oxidation treatments. Dibianca (1973), after viewing many hairs under the SEM, found

that the damage areas of cuticles, could generally be classified into four categories, i.e.,

fly away fibres, exposed cortex, split end, and general shaft damage.

With the “Knot Test” method, the knotted hair has a tendency of lifting on the cuticle

surface under high-stress bend; the higher the stress on the knot, the severer the hair

damage is, thus, the bigger extent of cuticle lifting. Figures 3.1-3.5 indicate features of

interest relevant to changes by the oxidation treatment on the single hair fibre’s surface

subjected to knot formation.

Untreated hairs in Figure 3.1A were examined in order to provide a standard visual

comparison. Bleaching treatments lead to degrading effects as shown in Figures 3.1B

and 3.2A. One of the obvious appearances in the bleached hairs is the increased

tendency of cuticle ‘lifting’ and this is possibly due to reduced adhesion of scales to

each other resulting from the solubilisation of CMC.

Figure 3.1: (A) Micrograph of an untreated hair fibre with somewhat broken edges; (B)

Micrograph of a bleached hair fibre. The appearance is somewhat similar to untreated

hair, though slightly greater uplift of scales is observed.

A B


51

Figure 3.2: (A) Micrograph of a bleached hair fibre, showing severe cuticle uplift; (B)

Micrograph of a bleached hair fibre, the greater lift up of scales is shown, and the hole

in the cuticle is observed.

Hair surfaces in Figure 3.2B and 3.3A show much more damage. A hole or rip is

apparent in the cuticle in Figure 3.2B with the greater lift up of scales. Parts of the

cuticle outer sublamellar layers were chipped away in Figure 3.3A and up-lifting of the

cuticle cells occurred. The sub-layers (A-layer, endocuticle, inner layer) are exposed.

The cortex is visible through the fracture.

The SEM image in Figure 3.3B demonstrates an almost complete lack of cuticle scales

on the fibre, where only some remnants of scales remain. Arrow indicates these

remnant cuticle scales. Probably, the cuticle of this fibre is more susceptible to removal.

Figure 3.3: (A) Micrograph of bleached hair fibre showing exposed cortex; (B)

Micrograph of bleached hair fibre, indicating a major lack of cuticle scales on the fibre,

with cuticle scales remnants remain.

A B

A B


52

The phenomena of cuticle loss or fragmentation can be observed in the Figure 3.4. The

surface of the hair has prominent transversal cracks in Figure 3.4A. The cuticle scales

are susceptible to separate from the hair, leading to exposed cortex by the extensive

cuticular detachment. Macrofibrils of the cortex are observed in Figure 3.4B through

the opening of the cuticle. Cortical macro- and micro-fibrils are clearly visible in the

fracture zone in Figure 3.5. Both step fracture and fragmentation are characteristic

patterns when the cuticle is in poor condition.

Swift (1972) showed that the cleavage of the cuticle cells generally occurs within the

endocuticle layer, so the fragments are observed. In this manner, the damaged fibre is

more susceptible to the oxidative degradation, thus accelerating the process of cuticle

degradation and scale removal.

Figure 3.4: (A) The fracture has exposed the underlying cortex; (B) Macrofibrils of

the cortex are observed through the opening of the cuticle.

Figure 3.5: Cortical macro-fibrils are clearly visible in the fracture zone of a bleached

hair fiber

Cortex

Macrofibri

l

A B


53

3.4.2 Knot Test

Cysteic acid derives from cystine by oxidative cleavage in hair fibre (Robbins, 1969).

Therefore, the destruction of the cross-link probably decreased the strength of the

cuticle and this decrease would cause scale lift and exfoliation of the cuticles during

the hair damage. For this reason, the “knot test” as a novel numerical description is

applied to evaluate the cuticle damage systematically. The damage of a hair knot is

assessed and classified based on two established standard damage class groups, as

shown in Figures 3.6 and 3.7. The acquired data for 30 fibres per hair tress and per

treatment is processed using statistic analysis in Excel software.

When the hair fibres are subject to bleaching damage, the cuticle is initially destroyed,

resulting in the cuticle edge being gradually degraded (Garcia, 1978). Figure 3.6 shows

a scheme which illustrates the progress of hair damage on mildly damaged fibres,

ranging from Damage Class A1 to A6. Generally, the lower the damage class, the less

is the extent of damage, thus less cuticle lifting occurs. For example, the individual

keratin scales in Class A1 have fairly smooth edges, and overlap, forming a flat, tight

sheath around the hair shaft (often called “virgin hair”), whereas Class A6

demonstrates complete cuticle lifting. In this way, observations on scale lifting provide

an effective method for distinguishing between the extents of damage.

Severe bleaching would cause the jagged-like edges of the cuticle scales to lift-up and

to become completely removed from the surface, exposing the underlying cortical

layers (Barton, 2011). As a consequence, a new damage classification for heavily

damaged fibres (Figure 3.7) is introduced to complement the damage classification

scheme.

According to the damage degree, there is a second group with six additional damage

classes, referred to Damage Classes B1 ─ B6, respectively. The total lack of surface

structure is seen in the most extreme case in Figure 3.7, where no cuticle scale or only

one layer of cuticle is left. Cracks start to appear on the last layer of cuticle in Class B2,

and pronounced transversal cracks are observed until this layer of cuticle chipped away,

thereby leaving the cortex exposed (Damage Class B6).


54

Figure 3.6: Standard damage classification for mildly damaged fibres


55

Figure 3.7: Damage classification for heavily damaged fibres

3.4.2.1 Results of the Knot-Test

The two groups of damage classifications, mildly damaged classification A and

heavily damaged classification B, were employed as a reference to be compared with

the target hair samples to achieve the corresponding numerical counts for each hair


56

sample. Hairs were classified three times into the various classes in order to achieve

reproducible results.

As described in the Chapter 2, there are two batches of virgin hair utilised in the

oxidative treatment, batch I and batch II. In Table 3.1 batch II virgin hair (N=30) has a

smaller fibre count in mildly damage group A2 (18.9%), A3 (28.9%) and A5 (7.8%)

than batch I virgin hair, which has 25.6%, 36.7% and 14.4%, respectively. However,

12.2% of batch II virgin hair (classified as heavily damage group B1) shows batch II

virgin hair has a slightly more sever damage extent than batch I virgin hair.

Table 3.1: Fibre count of virgin hair samples (N=30) in each classification based on

the mean of three measurements, as a percentage %. Standard error is given in the

Figure 3.8. Damage Class

Hair Batch

A1 A2 A3 A4 A5 A6 B1 B2 B3 B4 B5 B6

Batch I 5 25.6 36.7 20 14.4

Batch II 7.8 18.9 28.9 21.1 7.8 3.3 12.2

Figure 3.8 summarises the damage phenomena of each type of virgin hair in detail, the

main damage classifications of both virgin hairs belong to Damage A2, A3 and A4,

revealing that the virgin hairs are subjected to less damage attack. In addition, the

virgin hair batch II shows a relatively higher damage classification than the batch I,

this has been displayed in Table 3.1.

Figure 3.8: Fibre count of virgin hair samples (N=30) in each classification based on

the mean of three measurements, with standard error, as a percentage %

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

45.0%

A1 A2 A3 A4 A5 A6 B1

Fib

re C

ou

nt

in E

ach

Cla

ssif

icat

ion

(%

) N

=30

Damage Class

Virgin Hair Batch I

Virgin Hair Batch II


57

It is clear in Table 3.2 that during the various treatment times, most knot test results for

6% H202 bleached hairs are in mildly damage group A, except 3.3% of 1h 6% H202

bleached hairs are in heavily damage group B1. This indicates that 6% H202 bleach has

an overall moderate damage effect on the hair surface.

Table 3.2: Fibre count of 6% H202 bleached hair samples (N=30) in each classification,

at different bleaching times, based on the mean of three measurements, as a

percentage %. Standard error is given in the Figure 3.9.

A similar damage result is found in Figure 3.9 for the 6% H202 bleached hair. Most

damage classifications are found in damage classes A2, A3, and A4. In addition, there

are some other damage classifications in the mildly damage classification scheme A,

revealing that the 6% H202 bleached hairs have an overall moderate oxidative damage

effect.

Damage Class

Treatment Time

A1

A2

A3

A4

A5

A6

B1

B2

B3

B4

B5

B6

0.5 h 6.7 34.4 27.8 24.4 3.3 3.3

1h 13.3 22.2 37.8 16.7 3.3 3.3 3.3

1.5h 3.3 26.7 48.9 12.2 6.7 3.3

2h 14.4 35.6 40.0 10.0

2.5h 20.0 22.2 40.0 13.3 4.4

3h 20.0 25.6 26.7 17.7 6.7 3.3

3.5h 3.3 25.6 32.2 31.1 4.4 3.3

4h 4.4 21.1 37.8 21.1 8.9 6.7


58

Figure 3.9: Fibre count of 6% H202 bleached hair samples (N=30) in each

classification, at different bleaching times, based on the mean of three measurements,

with standard errors, as a percentage %

A distinctive result is observed for the 9% H202 commercially bleached hair in Table

3.3. 1.5h bleached hairs show milder damage effects, which are reflected in mild

damage classification A, whereas 9% H202 commercial bleached hairs after 2h shows

severer oxidised damage, which occurs in heavy damage classification B.

Table 3.3: Fibre count of 9% H202 commercial bleached hair samples (N=30) in each

classification, at different bleaching times, based on the mean of three measurements,

as a percentage %. Standard error is given in the Figure 3.10.

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

A1 A2 A3 A4 A5 A6 B1

Fib

re C

ou

nt

in E

ach

Cla

ssif

icat

ion

(%)

N=3

0

Damage Class

0.5h 1h 1.5h 2h

2.5h 3h 3.5h 4h

Damage Class

Treatment Time

A1 A2 A3 A4 A5 A6 B1 B2 B3 B4 B5 B6

0.5h 17.8 24.4 31.1 13.3 3.3 3.3 6.7

1h 5.6 28.9 23.3 32.2 6.7 3.3

1.5h 17.8 17.8 23.3 21.1 4.4 15.6

2h 37.8 32.2 13.3 10.0 6.7

2.5h 17.8 46.7 22.2 10.0 3.3

3h 22.2 31.1 26.7 14.4 5.6

3.5h 30.0 32.2 8.9 3.3 27.8

4h 52.2 12.2 11.1 3.3 5.6 16.7


59

The damage classifications for 9% commercially bleached hairs are observed in both

mild damage group A and heavy damage group B, as shown in Figure 3.10. As

described earlier, the results up to 1.5h fall into category A. The remaining bleached

samples show intensive damage and are classified in group B. The number of fibres in

classes B1 and B2 are largest, followed by A2, A3, and A4. Thus, 9% commercially

bleached hair is subjected to more sever oxidation effect on the cuticle than 6% H202

bleached hair.

Figure 3.10: Fibre count of 9% H202 commercial bleached hair samples (N=30) in

each classification, at different bleaching times, based on the mean of three

measurements, with standard errors, as a percentage %

The commercial persulphate bleached hairs display an unexpected phenomenon in

Table 3.3, their damages are mainly classified in Damage Class A and B1. 4h

commercial persulphate bleached hairs display 3.3% of fibres in damage classification

A1 and 90% of fibres in damage classification B1, indicates that the extended

treatment time results in the least topographical damage to the hair surface by the

commercial persulphate bleach.

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

A1 A2 A3 A4 A5 A6 B1 B2 B3 B4 B5 B6

Fib

re C

ou

nt

in E

ach

Cla

ssfi

cati

on

(%)

N=3

0

Damage Class

0.5h 1h 1.5h 2h

2.5h 3h 3.5h 4h


60

Table 3.4: Fibre count of commercial persulphate bleached hair samples (N=30) in


measurements, as a percentage %. Standard error is given in the Figure 3.11.

Commercial persulphate bleached hairs show a uniform damage effect in Figure 3.11.

Most of the bleached hairs are in Damage B1 with increasing treatment time. This

demonstrates that the extended commercial persulphate bleach reduced the damage

effect on the cuticle surface of the hair. However, further work will need to further

investigate this phenomenon.

Figure 3.11: Fibre count of commercial persulphate bleached hair samples (N=30) in


measurements, with standard errors, as a percentage %

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

90.0%

100.0%

A1 A2 A3 A4 A5 A6 B1 B2 B3 B4 B5 B6

Fib

re C

ou

nt

in E

ach

Cla

ssif

icat

ion

(%

) N

=30

Damage Class

0.5h 1h 1.5h 2h

2.5h 3h 3.5h 4h

Treatment

Time (h)

A1 A2 A3 A4 A5 A6 B1 B2 B3 B4 B5 B6

0.5 16.7 21.1 17.8 21.1 15.6 7.8

1 11.1 17.8 35.6 12.2 10.0 3.3 10.0

1.5 21.1 25.6 15.6 3.3 10.0 3.3 17.8 3.3

2 34.4 15.6 15.6 6.7 26.7 3.3

2.5 25.6 17.8 11.1 42.2 3.3

3 14.4 3.3 75.6 6.7

3.5 4.4 88.9 6.7

4 3.3 90.0 3.3 3.3


61

Due to the complexity of fibre count in each classification among the three bleaches,

an individual numerical damage classification is needed. The Equation 3.1 provides a

numerical specification of overall damage classification in each type of bleached hairs

(N=30) and the corresponding results are summarised in Table 3.5 - 3.9.

……… (3.1)

Where, Nn is the number of hair fibres in each group

According to the Equation 3.1, the fibre count of virgin hair samples (N=30) in each

classification is calculated in Table 3.5. It is apparent that both batch I and II virgin

hair have a similar damage group A value (~3.2), and the virgin hair type II has an

additional value in damage group B (0.9).

Table 3.5: Overall damage classification of two groups of damage classes for virgin

hair samples (N=30 fibres)

Hair Batches

Damage Group A

Damage Group B

Batch I

3.2

Batch II

3.2

0.9

The overall damage classification for the 6% H202 bleached hair is summarised in

Table 3.6. Comparatively, the uniform damage effect in damage group A appears, (as

reflected in damage class values of 2.5 ~ 3.3, except the 1h bleached hair, which has a

value of 1 in damage group B) to further prove that 6% H202 bleached hairs are

subjected to a moderate damage effect.

Overall Damage Classification =

+


62

Table 3.6: Overall damage classification of two groups of damage classes for 6% H202

bleached hair samples (N=30 fibres) at the different bleaching times

Treatment time (h) Damage Group A Damage Group B

0.5 2.9

1 2.8 1

1.5 3.1

2 2.5

2.5 2.6

3 2.8

3.5 3.2

4 3.3

The overall damage classification of the 9% H202 commercial bleached hair in Table

3.7 reveals that 1.5h of treatment has a slightly higher degree of damage, which is

around 2.7 ~ 3.2 for damage group A. However, longer bleaching times affect the

structural damage of the bleached hair after 2h, as observed in damage group B values

(approximately 2.3 ~ 3.0).

Table 3.7: Overall damage classification of two damage types for 9% H202

commercial bleached hair samples (N=30 fibres) at the different bleaching times


0.5 2.7 1

1 3.2

1.5 2.8 1.0

2 2.2

2.5 2.3

3 2.5

3.5 3.0

4 2.5


63

There is a gradual decrease from 3.0 to 1.0 in damage group A with the increasing

treatment time in Table 3.8, whereas the overall damage classification for the

commercial persulphate bleached hairs appears in damage group B as the value of 1.0

~ 1.9. The results show that the prolonged treatment time produces the least

topographical damage to the hair surface, thus, commercial persulphate bleach

treatment has an optimised oxidation effect on the hair cuticle with the extended

treatment time.

Table 3.8: Overall damage classification of two damage types for commercial

persulphate bleached hair samples (N=30 fibres) at the different bleaching times


0.5 2.9 1.0

1 3.0 1.0

1.5 2.5 1.9

2 2.0 1.1

2.5 1.8 1.1

3 1.3 1.1

3.5 1.3 1.1

4 1.0 1.3

3.5 Conclusion

SEM analysis of the surface morphology of hair knots showed that moderate to severe

damage was caused to the cuticle layer as a result of the various bleaching processes.

Such damage is more apparent in hair that had been bleached in 9% commercial H202

for 2hrs.

A systematic numerical approach to the “Knot Test” is applied to evaluate hair damage

situation based on two types of damage classifications: Mild Damage Classification A

and Heavy Damage Classification B. The results show that commercial persulphate

bleach treatment produced the least topographical damage with the prolonged

treatment time. 6% H202 bleach has an overall moderate damage effect over the


64

treatment time. 9% H202 commercial bleach displays an extensive damage result. The

first 1.5h bleached hairs are subject to the mildly oxidative treatment, and the damage

becomes more sever after 2 hours.

3.6 Reference

BARTON, P. M. J. 2011. A Forensic Investigation of Single Human Hair Fibres using

FTIR-ATR Spectroscopy and Chemometrics. Doctor of Philosphy (PhD),

Queensland University of Technology.

BROWN, A. C., SWIFT, J.A., 1974. Low voltage SEM of keratin fibre surfaces,.

Scanning Electron Microscopy, 241, 67-74.

CORCUFF, P., GREMILLET, P., JOURLIN, M., DUVAULT, Y., LEROY, F.,

LEVEQUE, J.L. 1993. 3D reconstruction of human hair by confocal

microscopy. J. Soc. Cosmet. Chem, 44, 1-12.

DALTON, J. S., ALLEN, G.C., HEARD, P.J., HALLAM, K.R., ELTON, N.J.,

WALKER, M.J., MATZ, G., 2000. Advancements in spectroscopic and

microscopic techniques for investigating the adsorption of conditioning

polymers onto human hair. Journal of Cosmetic Science, 51, 275-287.

DIBIANCA, S. P. 1973. Innovative Scanning Electron Microscopic Techniques for

Evaluating Hair Care Products. J. Soc. Cosmet. Chem, 24, 609-622.

FEUGHELMAN, M., WILLIS, B.K., 2001. Mechanical extension of human hair and

the movement of the cuticle. J Cosmet Sci., 52, 185-93.

GARCIA, M. L., EPPS, J. A., YARE, R. S., 1978. Normal cuticle-wear patterns in

human hair. J. Soc. Cosmet. Chem, 29, 155-175.

GILLEN, G., ROBERSON, S., NG, C., STRANICK, M., 1999. Elemental and

molecular imaging of human hair using secondary ion mass spectrometry.

Scanning. , 21, 173-81.

GURDEN, S. P., MONTEIRO, V.F., LONGO, E., FERREIRA, M.M.C., 2004.

Quantitative analysis and classification of AFM images of human hair. Journal

of Microscopy, 215, 13-23.

KADIKIS, A. 1987. Comments on "Quantitative Fiber Mixture Analysis by Scanning

Electron Microscopy". Textile Research Journal, 57, 676.

KREPLAK, L., MÉRIGOUX, C., BRIKI, F., FLOT, D., DOUCET, J., 2001.

Investigation of human hair structure by microdiffraction: direct observation of

cell membrane complex swelling. Biochim Biophys Acta., 1547, 268-74.


65

LANGLEY, K. D., KENNEDY, T. A., 1981. The Identification of Specialty Fibers.

Textile Research Journal, 51, 703.

MCMULLEN, R. L., KELTY, S. P., 2001. Investigation of Human Hair Fibers Using

Lateral Force Microscopy. Scanning., 23, 337-345.

PALENIK, S. 1983. Light microscopy of medullary micro-structure in hair

identification. Microscope, 31, 137.

POLETTI, G., ORSINI, F., LENARDI, C., BARBORINI, E., 2003. A comparative

study between AFM and SEM imaging on human scalp hair. Journal of

Microscopy, 211, 249-255.

ROBBINS, C. R. 1994. Chemical and Physical Behavior of Human Hair, Berlin,

Springer.

ROBBINS, C. R., KELLY, C., 1969. Amino acid analysis of cosmetically-altered hair.

J. Soc. Cosmet. Chem, 20, 555.

SICH, J. 1990. "Which is best for identifying animal fibers - Scanning electron

microscopy or light microscopy?" Textile chemist and colorist 22, 23-25.

SMITH, J. R. 1998. A quantitative method for analysing AFM images of the outer

surfaces of human hair. J. Microsc., 191, 223-228.

SWIFT, J. A. 1999. Human hair cuticle: biologically conspired to the owner’s

advantage. Journal of the Society of Cosmetic Chemists, 50, 23-47.

SWIFT, J. A., BROWN, A. C., 1972. The critical determination of fine changes in the

surface architecture of human hair due to cosmetic treatment Journal of the

Society of Cosmetic Chemists, 23, 695-702.

SWIFT, J. A., SMITH, J. R., 2000. Atomic force microscopy of human hair. Scanning.,

22, 310-318.

SWIFT, J. A., SMITH, J. R., 2001. Microscopical investigations on the epicuticle of

mammalian keratin fibres. Journal of Microscopy, 3, 203-211.

WORTMANN, F. J., ARNS, W., 1986. Quantitative fiber mixture analysis by

scanning electron microscopy: Part 1: Blends of mohair and cashmere with

sheep’s wool. Textile Research Journal, 56, 442-446.

Chapter 4 Colour Measurement of Human Hair

66


4.1 Introduction

Colour cannot exist without light, coloured light is perceived when the spectral

composition of the observed radiation varies over the visible range of wavelengths

(Hunt, 1995).

The electromagnetic radiation with wavelengths between 380 nm and 760 nm (790-

400 terahertz) is detected by the human eye and perceived as visible light. When a

beam of visible white light passed through a glass prism a series of coloured bands are

produced, resulting in seven colours ranging from violet to red and their wavelengths

are shown in Figure 4.1.

Figure 4.1: The visible portion of the spectrum is expanded to show the hues

associated with different wavelengths of light

When light strikes an object, the colour of a surface depends on how much light is

absorbed by the material and how much is scattered, transmitted and reflected at each

of the wavelengths in the visible spectrum (Berns, 2000). The specular reflection,

which is directly reflected at the surface of the fibre like a mirror, is not “coloured

light”.

4.1.1 Human Colour Vision & CIE Colour Space

The human eye is particularly sensitive to colour. It has been estimated that humans

can distinguish around 10 million colours (Wyszecki, 2006), thus, human colour vision

is introduced to consider the basic principles of colour perception.


67

4.1.1.1 Human Colour Vision

Different people may have different colour vision. Without the observers, who

perceive the light, there would be no colour (Hunt, 1998). So the eye plays a role of a

transducer in converting the electromagnetic energy, which is in the form of light, into

the nerve impulses (Hill, 1997).

Light entering our eyes is imaged onto the retina at the back of the eye. The eye

contains two types of light-sensitive cells, called rods and cones. They both are located

in the retina in the last layer of nerve cells furthest from the cornea. Rods are not

colour-sensitive cells, they work with the weak illumination (scotopic or night vision),

but cones detect light with high illumination (photopic vision). There are three types of

cone receptors with a maximum sensitivity for wavelengths in the red, green and blue

regions of the visible spectrum, called short-wavelength-sensitive (S) cones, middle-

wavelength-sensitive (M) cones, and long-wavelength-sensitive (L) cones, respectively.

Thus, in principle, three primaries (R, G, B) describe a colour sensation. If either of the

cones is missing or defective, it causes the defective colour vision.

As illustrated in the Figure 4.2, the spectral sensitivity as a function of wavelength,

varies depending on the type of cone. S- cones are sensitive to the short wavelength

region, M- and L- cones are sensitive to the middle and long wavelength regions,

respectively. Particularly the L-cone and the M-cone spectral sensitivities are close

together compared to the S-cone sensitivity, thus improving colour discrimination

(Berns, 2000).

Figure 4.2: Normalized response spectra of human eye cones, S, M, and L types, to

monochromatic spectral stimuli with wavelength in nanometres (Stockmam, 1993).


68

The human observer perceives colour as a ratio of the intensities from the S, M and L

cones. Colour perception starts with the spectral characteristics of the light source,

which are then modified by the reflectance of the object (Battle, 1997). The resulting

light then stimulates the eye to generate a signal, which is eventually interpreted by the

brain.

4.1.1.2 CIE system

Colour perception always depends on three elements: the light, the object and the

observer. Colour is a complex perception, it changes under different sets of conditions,

such as source (e.g., spectral properties and the amount of light), object (e.g., size and

texture) and observer (e.g., spectral sensitivities, lens properties, macular pigment,

light and chromatic adaptation) (Berns, 2000). To measure the colour of an object

objectively, the international committee (Commision Internationale de L’Eclairage,

CIE) set up a standardised colour specification system in 1931, refer as 1931 CIE

system. Although additional systems have been added, the basic principles and

systems are the same. And almost all modern colour measurements are based on the

CIE system.

4.1.1.2.1 The 1931 CIE system

In 1931 the CIE developed a system for specifying colour stimuli using tristimulus

values for three imaginary primaries, the amounts of the three primaries are called

tristimulus values and are referred as X, Y, and Z in the 1931 CIE system. Tristimulus

values can be calculated by the reflectance spectrum of a sample, and the reflectance

spectrum can be measured using a reflectance spectrophotometer.

Tristimulus values provide a numerical specification of the colour. A different colour

has different tristimulus values and hence the specification is different (Rigg, 2007). In

order to eliminate the variations of colour specification, the CIE had defined standard

light sources and a standard observer, together with standard observing and viewing

conditions.


69

Standard illuminants and Standard sources

The most important light source is daylight. A light source can be quantified by

measuring its spectral power distribution (SPD), which is a function of wavelength

across the visible spectrum. CIE makes a distinction between sources and illuminants.

A light source is a physical emitter of light, such as the sun or a lamp, an illuminant

which refers to a specified spectral energy distribution, is the specification for a

potential light source. All light sources can be specified as an illuminant, but not all

illuminants can be physically realised as a light source.

In 1931, the CIE recommended three standard illuminants, known as A, B and C,

respectively. llluminant A represents incandescent light of a tungsten filament lamp,

which has the same SPD as a Planckian or black body radiator at a colour temperature

of 2856 K. A black body is an ideal thermal radiator. The SPD of the emitted light

from a black body depends only on the temperature of the surface, but not its nature.

The higher the temperature of the emitting surface, the greater the total power of the

emitted radiation, and the lower the wavelength of maximum emission (Broadbent,

2001). The temperature of a black body is measured from its colour temperature,

usually expressed in Kelvins (K).

The CIE also defined illuminant B: tungsten with a yellow filter, which simulates

sunlight at the correlated colour temperature of 4874K, and illuminant C: tungsten

with a blue filter, which simulates daylight at the correlated colour temperature of

6774K, however, both illuminants are not suggested.

In 1963, the CIE recommended a series of D illuminants based on natural daylight

across the UV, visible and near–IR region (300-830 nm) (Sinclair, 1997). It enables

daylight spectral power distributions to be calculated for a wide range of correlated

colour temperatures. CIE illuminant D65, with an approximate correlated colour

temperature of 6500K, is now accepted as a standard illuminant (CIE 1986) for surface

colour industries, and D50 is used in the graphic arts and computer industries (Berns,

2000).


70

Standard Observers

The original 1931 CIE standard observer was based on colour-matching function using

a 2o

field of view, whose visual field size is between 1º and 4º. Thus the CIE 1931

standard observer is also known as the CIE 1931 2° standard observer. But a 2o field of

view is much narrower than that normally used for the critical colour appraisal (Rigg,

2007), and cannot remain a match for large-field visual colour judgements. In 1964 the

CIE adapted a new standard observer to describe the vision of a human more

accurately, which is called 10o observer or 1964 CIE supplementary standard observer.

Although neither is likely to correspond closely to the individual observer, either may

well correspond reasonably closely to a real observer (Rigg, 2007).

The standard observer is characterised by colour-matching functions. Colour-matching

functions define how human eyes match a colour stimulus with an additive mixture of

three primaries: monochromatic red, green and blue lights. According to the

trichromatic theory of colour vision, any colour stimulus can be specified by the

amounts of the primaries to match a particular colour.

Viewing condition

The CIE has recommended four types of illumination and viewing geometries for

reflectance measurement: 45º/normal (45/0), normal/45º (0/45), normal/diffuse (0/d)

and diffuse/normal (d/0), as presented in the Figure 4.3.

In the 45/0 geometry, the sample is illuminated with one or more beams of light at the

incident angle of 45º to the surface and the light is normally viewed. In the opposite

mode (0/45 geometry), the sample is illuminated normally to its surface and measures

at about 45º angle, while the incident light is polarised. The 45/0 and 0/45 geometries

both ensure all components of gloss to be excluded from measurements.

In the 0/d geometry, the sample is illuminated from an angle near the normal and the

reflected energy is collected from all angles using an integrating sphere (a hollow

sphere which is painted white inside). In the d/0 geometry, the sample is illuminated


71

from all angles using an integrating sphere and viewed at an angle near the normal to

the surface. These two geometries are optically reverse to each other and therefore

result in the same measurement results.

But most D/0 instruments are actually using D/8 rather than true D/0 geometry, a small

amount of specular component may be excluded by a dark spot or light trap on the

sphere, thus sphere instruments may allow specular reflection to be included or

excluded (Figure 4.4). In the specular-included mode, the total reflectance (diffuse +

specular) can be measured.

Figure 4.3: CIE-recommended illuminating and viewing geometries (Battle, 1997)

Figure 4.4: Comparison between D/8 specular-included and specular-excluded modes

(Battle, 1997)


72

4.1.1.2.2 CIE 1976 Colour system

As mentioned previously for the colour-matching function, it is difficult to relate the

tristimulus values of an object to its appearance because the XYZ values have very

little real meaning to most observers (Rigg, 2007). Consequently, a two-dimensional

plane with the lightness (Y) of a sample was introduced by projecting the tristimulus

values onto the unit plane (x+y+z=1), this plane is known as the chromaticity diagram

(CIE xyY), and its chromaticity coordinates are recorded as x and y (Klaman, 2002).

However the CIE xyY colour space is not perceptually uniform and the colour

differences in the colour space are not uniform, which means equal distances in the

diagram do not correspond to equal visual differences (Hunt, 1971, Ohno, 2000).

Therefore, after a number of non-linear transformations of the CIE 1931 XYZ space, a

uniform colour space, namely CIE 1976 LAB (or CIE L*a*b*) colour space as one of

these transformations was finally introduced. A three-dimensional plot of the

tristimulus values XYZ determines a colour space.

CIE L*a*b*, or CIELAB system is a very popular colour scale for colour and colour

difference specification. It measures colour on three axes that are nearly linear with

human perception (Ford, 1998). This model has the benefit of corresponding to the

human perception of colour and provides a grid point for each specific colour (TASI,

2004).

In the CIE L*a*b* colour system, the lightness (or intensity) of a colour is measured

on the vertical L* axis. The maximum for L* is 100, which represents a perfect

reflecting diffuser. The minimum for L* is zero, which represents black. The a* and b*

axes have no specific numerical limits. Positive a* is red. Negative a* is green.

Positive b* is yellow. Negative b* is blue. Below a diagram representing the CIELAB

colour space is shown in Figure 4.5. One unit on the L*, a*, or b* axes is considered to

be the limit of human discrimination between colour (TASI, 2004). This colour grid

allows the mathematical comparison of colours and also allows colours to be corrected

for different conditions. Our research has used the cut-off point of ±100 because there

are the practical limits of the software and instrument used in the colour measurement

(Napier, 2007).


73

Figure 4.5: CIE L*a*b* colour system (Qian, 2012)

Colour Difference

Two non-matched colours always have different sets of the tristimulus values X, Y and

Z. Since the CIE XYZ colour space is not a uniform colour space, it is not particularly

useful for colour difference evaluation (Broadbent, 2001). Consequently, the CIE LAB

colour difference equation is used for the colour specification.

A colour-difference space is a three-dimensional space, the total colour difference

between two non-matching colours is the Euclidean distance between the points for

their respective coordinates in the CIELAB colour space (Broadbent, 2001), given by:

ΔE* = [(ΔL*) 2

+ (Δa*) 2

+ (Δb*) 2

]1/2

(4.1)

Where ΔL* refers to the lightness difference (lighter if positive, darker if negative),

Δa* refers to the red-green difference (redder if positive, greener if negative) and Δb*

refers to the yellow-blue difference (yellowish if positive, bluer if negative).

The value, ΔE*, defines the extent of the total colour difference, but it does not

provide information about how the colours differ.


74

4.1.2 Colour Measurement

There are two types of instrument to measure the colour of opaque samples:

reflectance spectrophotometers and colorimeters.

4.1.2.1 Reflectance Spectrophotometry

Reflectance spectrophotometry is an established method for measuring hair colour at

the macroscopic level, and it has been more consistent than digital image analysis

(Vaughn, 2008, 2009).

Spectrophotometric measurements provide the numerical description of the reflection

or transmission of light by an object. The resulting reflection or transmission spectrum

gives the fraction of the incident light that an object reflects or transmits as a function

of wavelength (Broadbent, 2001). The reflectance measures the ratio of reflected to

incident light from a sample at many points across the visible spectrum (see in Eqn

4.2). The reflectance of a sample is generally expressed as percentage, and the perfect

reflecting diffuser has a reflectance of 100%. But the obtained reflectance values are

the relative values. The light used to illuminate the sample does not affect reflectance

and transmittance. Thus, XYZ, or L*a*b* coordinates can be calculated with the

illuminant (Battle, 1997).

Reflectance =

(4.2)

It has to be noted that there is no light transmission through the material and only

reflection from the opaque sample surface. Figure 4.6 illustrates a schematic of a

reflection spectrophotometer. A light source is used to illuminate the sample using a

specific illumination and viewing geometry. Reflected or transmitted light is then

passed to the spectral analyser, where the light is resolved into its spectral components.

This allows the light detector and control electronics to be measured across the visible

spectrum. And a highly diffusive, reflecting, white coating covers the interior wall of

the sphere to achieve accurate diffuse measurements.


75

Figure 4.6: Schematic of a reflection spectrophotometer with an integrating sphere

(Broadbent, 2001). The incoming light beam is diffusely reflected inside an integrating

sphere, when using the D/8 geometry, some of the light, which comes from specular

reflection from sample, then passing to the detector.

4.1.2.2 Colorimeters

A tristimulus colorimeter is a very cost-effective and portable colour-measuring

instrument and has the benefit of being able to measure colour difference in a quality-

control situation (Battle, 1997). As the human eye, it has red, green and blue photo

detectors, which measure tristimulus values of the light (Figure 4.7). When the light

source, such as a quartz halogen bulb, illuminates the sample at 45o to the normal, the

reflected light is collected and then passed to a detector, which consists of three filters

in front of its own light-sensitive diode. Thus, the response of the filter/diode

combination corresponds to the differential spectral response of the eye (Battle, 1997).

But its absolute accuracy is restricted, since the practical manipulation of the light

source responses to the filter/diode detector always just satisfies the CIE definitions.

The illuminant and observer data are fixed as D65 and 10o, and the instrument is

restricted to determine the relative differences (Malacara, 2002). As a result, a

colorimeter cannot measure the metamerism. Metamerism occurs when two samples

with different spectral reflectance data may look the same with a given light source,

but are very different under another.


76

Figure 4.7: The basic features of a colorimeter with 45/0 geometry (McDonald, 1997)

4.2 Objectives

It has been verified that instrumental assessment determines colour differences more

accurate than average human observers. The purpose of this colour investigation was

to see whether a colour-measuring unit could provide a reproducible, objective method

in determining the colour changes of bleached human hairs utilizing the CIE L*a*b*

colour space. Three types of bleached hairs were used, namely, 6% H202 bleach, 9%

H202 commercial bleach and commercial persulphate bleach.

4.3 Experimental

Hair colour was measured using a reflectance spectrophotometer (Spectraflash 600,

Datacolor, Switzerland) with D/8 viewing geometry, 10o CIE standard observer and

D65 light source, which represents daylight without spectral highlights.

Before each measurement the instrument was calibrated using a standard white tile

(supplied with the machine), whose reflectance at each wavelength is known compared

to a perfect diffuse reflecting surface.


77

In the colour measurement, the samples of virgin and bleached hairs were examined.

Due to the non-uniformity of the hair, it is important to examine the variation along the

length on each hair tress. Hair samples were twisted tightly and placed vertically with a

special holder against the measuring aperture of the instrument. They were measured

at four randomly chosen points on each hair tress in the direction of tip to root. Three

more measurements were performed on the same hair tress from left, right and back to

assess the reproducibility of the readings. Three tresses per treatment were tested to

obtain a representative average value. Although the differences arising from variations

of hair samples exist, they were found to be negligible compared with those arising

from the treatment given to the hair.

Reflectance spectrophotometer measures the reflected wavelengths of the incident light

and their intensity of the hair samples. Measured data were imported into the Microsoft

Excel 2007 for calculating the values of the colour parameters. Irrelevant details refer

to Chapter 8: Appendix.

4.4 Results and Discussion

The pigments contained in hair consume hydrogen peroxide at a measurably faster rate

than hair without pigment (Wolfram, 1970a, Hall, 1975). Thus peroxide reacts faster

with hair pigment than with hair protein (Robbins, 2002), so that the initial response of

oxidative change to hair fibre is colour change.

The raw data of L* (lightness), a* (red - green colour axis), b* (yellow - blue colour

axis) of the three types of bleached hair are shown in Tables 4.1, 4.2 and 4.3,

respectively. Since there are no negative L*, a* and b* values in the tables, none of the

bleached hair has a blue or green or darker hue. All bleached hair samples have an

overall brighter colour change, as was expected.

In Table 4.1, 6% H202 bleached hairs display an increase in L* value from 27.6 to 50.1

and b* value from 9.8 to 26.4, while the a* value shows only a relatively minor change

from 5.7 to 9.0. However the a* value behave differently in Tables 4.2 and 4.3, where

L* and b* values have a considerably increase over the treatment time. L* values


78

increase from 27.6 to 62.5 for 9% H202 commercial bleach and 28.7 to 75.3 in

commercial persulphate bleach. Similarly b* values increase from 9.8 and 10.8 to 20.3

and 23.9 for 9% H202 commercial bleach and commercial persulphate bleach,

respectively.

Table 4.1: The CIELAB results for 6% H2O2 bleached hair samples at different

bleaching times, in the form of the arithmetic means with standard error

Bleaching Time (h) a* b* L*

0 5.7±0.0 9.8±0.1 27.6±0.2

½ 8.5±0.2 17.2±0.2 34.1±0.3

1 9.5±0.0 21.5±0.1 39.7±0.0

1½ 9.6±0.1 22.8±0.1 41.7±0.3

2 9.7±0.1 24.7±0.1 44.8±0.2

2½ 9.6±0.0 24.3±0.2 44.8±0.1

3 9.4±0.1 24.9±0.2 45.9±0.3

3½ 9.0±0.0 24.9±0.0 47.4±0.2

4 9.0±0.1 26.4±0.1 50.1±0.2

Table 4.2: The CIELAB results for 9% H2O2 commercial bleached hair samples at

different bleaching times, in the form of the arithmetic means with standard error


0 5.7±0.0 9.8±0.1 27.6±0.2

½ 9.7±0.1 22.5±0.3 40.8±0.4

1 9.3±0.1 24.3±0.4 45.3±0.9

1½ 8.9±0.1 25.6±0.2 49.8±0.3

2 7.7±0.3 24.7±0.3 54.1±0.9

2½ 6.3±0.1 23.1±0.3 57.5±0.7

3 6.8±0.2 23.6±0.2 56.1±0.7

3½ 5.3±0.2 21.3±0.2 62.3±0.6

4 5.0±0.1 20.3±0.1 62.5±0.8


79

Table 4.3: The CIELAB results for commercial persulphate bleached hair samples at

different bleaching times, in the form of the arithmetic means with standard error


0 6.4±0.1 10.8±0.2 28.7±0.4

½ 10.2±0.1 27.4±0.1 49.9±0.3

1 9.5±0.1 29.0±0.1 55.1±0.2

1½ 8.6±0.1 30.0±0.2 60.1±0.2

2 7.2±0.1 29.3±0.2 65.2±0.3

2½ 6.6±0.1 28.9±0.1 67.2±0.2

3 5.1±0.1 27.1±0.2 70.8±0.2

3½ 3.4±0.1 25.4±0.3 73.4±0.2

4 2.8±0.1 23.9±0.2 75.3±0.3

The colour difference parameters: ∆L*, ∆a*, ∆b* and ∆E* were calculated according

to the Equation 4.1, and the corresponding results are given in Tables 4.4, 4.5 and 4.6.

The values of ∆L*, ∆b* and ∆E* for the bleached hairs have positive values, which

means that bleached hair has a tendency of being lighter and more yellow in

comparison to the reference virgin hair. Although some ∆a* values are negative,

bleached hair still has a redness hue, but the value is smaller than for virgin hair.

Table 4.4: The colour changes of ∆L*, ∆a*, ∆b* and ∆E* for 6% H2O2 bleached hair

samples at different bleaching times

Bleaching Time (h) ∆a* ∆b* ∆L* ∆E*

0 ---- ---- ---- ----

½ 2.8 7.4 6.4 10.2

1 3.8 11.7 12.1 17.2

1½ 3.9 12.9 14.1 19.5

2 3.9 14.8 17.2 23.1

2½ 3.9 14.4 17.2 22.8

3 3.7 15.0 18.2 23.9

3½ 3.3 15.0 19.8 25.1

4 3.3 16.5 22.5 28.1


80

Table 4.5: The colour changes of ∆L*, ∆a*, ∆b* and ∆E* for 9% H2O2 commercial

bleached hair samples at different bleaching times


0 ---- ---- ---- ----

½ 3.9 12.7 13.2 18.7

1 3.6 14.4 17.7 23.1

1½ 3.1 15.8 22.2 27.5

2 1.9 14.9 26.5 30.5

2½ 0.6 13.3 29.9 32.7

3 1.1 13.8 28.5 31.7

3½ -0.4 11.5 34.7 36.5

4 -0.7 10.9 34.9 36.5

Table 4.6: The colour changes of ∆L*, ∆a*, ∆b* and ∆E* for commercial persulphate

bleached hair samples at different bleaching times


0 ---- ---- ---- ----

½ 3.8 16.5 21.2 27.1

1 3.1 18.2 26.5 32.3

1½ 2.2 19.1 31.5 36.9

2 0.7 18.5 36.5 41.0

2½ 0.2 18.1 38.5 42.6

3 -1.4 16.3 42.2 45.2

3½ -3.0 14.5 44.7 47.1

4 -3.6 13.0 46.6 48.5

Bleached hair fibres are subjected to an oxidative degradation effect, which impacts on

the melanin granules, leading to colour loss. The quantity and the distribution of

eumelanin and pheomelanin in hair varies in their ratio (Wolfram, 1987), so that a*-

and b*- values were used to examine the decomposition of the pigments in bleached

hair. The graphs of 9% H2O2 commercial bleach and commercial persulphate bleach


81

exhibit a similar trend in Figures 4.8 and 4.9. They both increase at the beginning and

then decrease towards the end of the bleaching time.

As mentioned in Chapter 1, bleaching is a diffusion-controlled process (Robbins,

2002). a*-and b*- values indicate a significant colour change at the beginning of the

reaction time. Even at the early stage, the peroxide has made its way into the cortex to

oxidise the melanins. The time for a* to reach its maximum value (0.5h) is shorter than

for b* (1.5h). This is due to the fact that red pheomelanin is more resistant to oxidative

degradation than the black-brown eumelanins, hence, eumelanins are oxidised earlier

than pheomelanins. A sudden drop of both a*- and b*- values suggests that the

oxidation of pheomelanin starts after 0.5h and the degradation of protein occurs once

the eumelanins were dissolved (1.5h), which leads to the reduced b*- values. The

black-brown eumelanins are the main colour contributors for virgin Caucasian hair.

The final yellow colour for bleached hair is ascribed to the natural colour of hair

keratin (Wolfram, 1970a).

As a result, hair samples show a progressive damage effect after attacked by 9% H2O2

commercial bleach, and especially by the commercial persulphate bleach, where the

bleaching effect is enhanced in the presence of the persulphate with the peroxide

(Robbins, 2002).

On the contrary, the values of a* and b* for 6% H2O2 bleached hairs show a stable

trendline relationship in Figures 4.8 and 4.9. The values keep increasing to achieve

their equilibrium level during the whole treatment, revealing that 6% H2O2 bleach has

a mild degradation effect on the melanin of the hair. It is still decolorising the

eumelanin at this stage, while the other two bleaches start oxidising the keratin. This

conclusion confirms the observations from the SEM “Knot Test” investigations that 6%

H202 has an overall moderate damage effect over the oxidation time.


82

Figure 4.8: Values of a* for 3 bleaches: (6% H2O2 bleach, 9% H2O2 commercial

bleach, commercial persulphate bleach) for up to 4 hours bleaching time. The trendline

curves are empirical and based on polynomial relationships.

Figure 4.9: Values of b* for 3 bleaches: (6% H2O2 bleach, 9% H2O2 commercial



Three treatments have a consistent upwards tendency in L*, as shown in Figure 4.10.

Melanin pigments are oxidised by peroxide bleach, so the hair samples become lighter.

Commercial persulphate bleach exhibits a larger lightness value than 9% H2O2

commercial bleach and 6% H2O2 bleach, which shows that the stronger oxidising agent

(persulphate) in the commercial persulphate bleach gives rise to a stronger degradation

0

2

4

6

8

10

12

0 1 2 3 4

a*

Bleaching Time (h)

6% H202 Bleach

9% H202 Commercial Bleach

Commercial Persulphate Bleach

0

5

10

15

20

25

30

35

0 1 2 3 4

b*

Bleaching Time (h)

6% H202 Bleach




83

of the pigment and thus to a large increase in the lightness value. However, the three

types of bleach haven’t reached their equilbrium level during the whole treatment time

up to 4 hours.

Figure 4.10: Values of L* for 3 bleaches: (6% H2O2 bleach, 9% H2O2 commercial



The total colour difference ∆E* of three bleaches are calculated by Eqn.4.1 in Tables

4.1, 4.2 and 4.3, and are presented in Figure 4.11. ∆E* is a single value which takes

into account the differences between L*, a* and b* of the bleached hair and standard

virgin hair. It may potentially be misleading in some cases where ∆L*, ∆a* and ∆b*

provide negative values, when ∆E* is still within the tolerance.

In Figure 4.11 it shows that the progressive change of hair colour occurs with

increasing bleaching time. An obvious colour change is observed for the strongest

bleach and the longest bleaching time. The higher the concentration of peroxide, the

longer the treatment time, the larger is the ∆E*-value, thereby the greater is the hair

colour difference. Commercial persulphate bleach thus leads to an overall larger colour

change than 9% H202 commercial bleach and 6% H202 bleach. However, there is no

equilibrium for ∆E*- value seems to be reached for three types of bleach.

0

10

20

30

40

50

60

70

80

0 1 2 3 4

L*

Bleaching Time (h)

6% H202 Bleach




84

Figure 4.11: Values of ∆E* for 3 bleaches: (6% H2O2 bleach, 9% H2O2 commercial



The redness hue is caused by the breakdown and the solubilisation of the pigment

particles in the cortex (Wolfram, 1987) , but it has been suggested that the a* of hair

plays very little role in determining the colour change. Thus b* and L* are the main

components in the CIE L*a*b* colour evaluation.

The changes of L* and b* occur with the melanin oxidation process. A consistent

increase in L* as well as in b* over the reaction time is observed on 6% H202 bleach in

Figure 4.12, giving evidence that they both have a homogenous damage effect on the

melanin. An up-turn of the curves seem to appear in the last half hour, which may be

attributed to the considerable solubilisation of the black-brown melanin granules in

this mild oxidation treatment.

0

10

20

30

40

50

0 1 2 3 4

∆E

*

Bleaching Time (h)

6% H202 Bleach




85

Figure 4.12: The relationship of L*- and b*- values for 6% H202 bleach over the 4h

bleaching time. The trendline curves are empirical and based on polynomial

relationships.

It is worth pointing out that the plots of L* vs. b* in Figures 4.13 and 4.14 for 9% H202

commercial bleach and commercial persulphate bleach follow a similar trend. b*-

values increases initially and then decrease to interact with the L*- curve, which keeps

the upwards tendency all the time. This shows that the solubilisation of eumelanins is

the primary prerequisite in determining the change of b*. The reduced b*- values result

from the oxidation of keratin. L* would have a dramatic change when the melanin has

been oxidised.

The two curves meet at 3h and 2.5h for 9% H202 commercial bleach and commercial

persulphate bleach, respectively. The reason for the earlier meeting point for

commercial persulphate bleach is because this type of oxidised hair has a relatively

higher L* and b* values, caused by an aggressive oxidising agent, such as persulphate.

0

5

10

15

20

25

30

35

20

30

40

50

60

70

80

0 1 2 3 4

b*

L*

Bleaching Time (h)

Lightness Yellowness


86

Figure 4.13: The relationship of L*- and b*- values for 9% H202 commercial bleach

over the 4h bleaching time. The trendline curves are empirical and based on

polynomial relationships.

Figure 4.14: The relationship of L*- and b*- values for commercial persulphate bleach

over the 4h bleaching time. The trendline curves are empirical and based on

polynomial relationships.

Oxidised melanin, causing the changes in L* and b* for 9% H202 commercial bleach

and commercial persulphate bleach, display a similar semi- circular shape of the graph

in Figure 4.15. They both increase in a regular pattern until reaching their maximum

values, where the curves start decreasing, and these maximum values appear when

0

5

10

15

20

25

30

35

20

30

40

50

60

70

80

0 1 2 3 4

b*

L*

Bleaching Time (h)


0

5

10

15

20

25

30

35

20

30

40

50

60

70

80

0 1 2 3 4

b*

L*

Bleaching Time (h)



87

black-brown melanin had been completely degraded and dissolved, as described in the

Figures 4.8 and 4.9.

6% H202 bleach curve shows a uniform increase for both b*- and L*- values. b* and

L* increase simultaneously, which demonstrates that 6% H202 bleached hair has a

homogenous damage effect on the yellowness and lightness. But it overlaps with some

of the 9% H202 commercial bleach values. This provides an indication that 6% H202

bleach would decrease with further bleaching time once it reaches the maximum value,

and this further proves that the 6% H202 bleach has a mild damage effect on the

melanin granules. In comparison, commercial persulphate bleach has larger values of

L* and b*, because it is the stronger oxidising agent.

Figure 4.15: The plot of L* vs. b* for hairs subjected to three types of bleaches

The spectral reflectance character of the three types of bleach is used as a guide to the

effects of bleaching on the colour characteristics of the hair in the visible region in the

Figures 4.16, 4.17, and 4.18, respectively.

Overall, there is a dominant increase in the wavelength on the three bleaches with the

prolonged treatment times, shown in Figures 4.16, 4.17, and 4.18. The longer

bleaching times lead to a higher reflectance of light. On the other hand, different

bleaches result in different reflectance curves. Commercial persulphate bleach shows

the largest reflectance, followed by 9% H202 commercial bleach and 6% H202 bleach.

0

5

10

15

20

25

30

35

20 30 40 50 60 70 80

b*

L*

6% H202 Bleach




88

It is well known that when light strikes the hair, most of the light is absorbed in dark

coloured hair, thus leaving less light reflected. This phenomena provides the evidence

that commercial persulphate bleached hair has a relatively lighter appearance than 9%

H202 commercial bleach and 6% H202 bleach.

The overall reflectance values for 6% H202 bleach are presented in Figure 4.16. The

curve increases significantly in the first 1h reaction time, and then gradually

approaches its maximum value. The significant change can be understood in that

bleaching is a diffusion-controlled process and the steady increase occurs after

overcoming the initial slow rate of diffusion (Hessefort, 2008).

Figure 4.16: Reflectance of 6% H202 bleached hair samples in the visible spectrum.

The raw data is given in Chapter 8, Table 8.1.

Relatively, commercial persulphate bleached hair behaves oppositely. The curves in

Figure 4.17 observe a major change for 0.5h. Each curve shows a uniform increase for

the whole region, with even spacing. This phenomenon indicates that this type of

bleaching leads to an overall larger colour change and the bleached hair becomes much

lighter over time. This conclusion is also reflected in Table 4.3.

0

10

20

30

40

50

60

70

400 450 500 550 600 650 700

% R

efle

ctan

ce

Wavelength (nm)

Original Hair 0.5h

1h 1.5h

2h 2.5h

3h 3.5h

4h

Increasing

Bleaching

Time


89

Figure 4.17: Reflectance of commercial persulphate bleached hair samples in the

visible spectrum. The raw data is given in Chapter 8, Table 8.2.

9% H202 commercial bleach demonstrates some similarities to both 6% H202 bleach

and commercial persulphate bleach in Figure 4.18. The curves show a large difference

for 0.5h, and equal spacing of the curves for the rest of the reaction time. In addition,

the curves for 2.5h and 3h, and the curves for 3.5h and 4h overlap. This is attributed to

the similar reflectance of the bleached hair surfaces.

Figure 4.18: Reflectance of 9% H202 commercial bleached hair samples in the visible

spectrum. The raw data is given in Chapter 8, Table 8.3.

0

10

20

30

40

50

60

70

400 450 500 550 600 650 700

% R

efl

ect

ance

Wavelength (nm)

Original Hair 0.5h

1h 1.5h

2h 2.5h

3h 3.5h

4h

0

10

20

30

40

50

60

70

400 450 500 550 600 650 700

% R

efle

ctan

ce

Wavelength (nm)

Original Hair 0.5h

1h 1.5h

2h 2.5h

3h 3.5h

4h Increasing

Bleaching

Time

Increasing

Bleaching

Time


90

4.5 Conclusion

It is known that the initial response of oxidative damage of a hair fibre is the colour

change. The CIE LAB colour space as an effective tool for the study of colour

specification was employed to evaluate the damage effects caused by the three

bleaches.

In general, the bleached hairs have an overall lighter, yellowish and reddish colour. It

is obvious that the lightness of the bleached hair is associated with pigment oxidation,

whereas the yellowness value is determined by two steps. The maximum value appears

when the solubilisation of black-brown melanin has been accomplished and is

followed by a reduced value as a result of protein (keratin) degradation.

Since the a*- value of hair plays very little role in determining the colour change, the

b*- and L*- values are the main contributors in the CIE L*a*b* colour evaluation. The

higher the concentration of peroxide, the longer treatment time, the larger is the ∆E*-

value, thereby the greater is the hair colour difference. 9% H202 commercial bleach and

especially the commercial persulphate bleach demonstrate a higher overall increase in

colour change than the 6% H202 bleach, which only has a mild oxidation effect.

Consequently, the commercial persulphate bleached hair is much lighter and more

yellow than 9% commercial bleached hair and 6% bleached hair.

4.6 References

BATTLE, D. R. 1997. The measurement of colour. In: MCDONALD, R. (ed.) Colour

Physics for Industry. Second Edition ed. Bradford, England: Society of Dyers

and Colourists.

BERNS, R. S. 2000. Billmeyer and Saltzman's principles of color technology, New

York, A Wiley-Interscience Publication.

BROADBENT, A. D. 2001. Basic Principles of Textile Coloration, Society of Dyers

and Colourists.


91

FORD, A., ROBERTS, A., 1998. Colour Space Conversions. London: Westminster

University, .

HALL, K., WOLFRAM, L.J., 1975. Isolation and identification of the hair protein

component of hair melanin. Journal Cosmetic Chemists, 26, 247.

HESSEFORT, Y., HOLLAND, B.T., CLOUD, R.W., 2008. True porosity

measurement of hair: a new way to study hair damage mechanisms. J Cosmet

Sci,, 59, 303-315.

HILL, A. R. 1997. How we see colour. In: MCDONALD, R. (ed.) Colour Physicas for

Industry. Second Edition ed. Bradford, England: Society of Dyers and Colorists.

HUNT, R. W. G. 1971. Colour Measurement Review of Progress in Coloration 2, 11-

19.

HUNT, R. W. G. 1995. The reproduction of colour in Photography, Printing &

Televison, Kingston-upon-Thames, Fountain press.

HUNT, R. W. G. 1998. Measuring Colour Kingston-upon-Thames, Fountain Press.

KLAMAN, M. 2002. Aspects on Colour Rendering, Colour Prediction and Colour

Control in Printed Media. Royal Institutes of Technology.

MALACARA, D. 2002. Colour Vision and Colorimetry: Theory and Application,

Washington, SPIE-The international Society for Optical Engnieering.

MCDONALD, R. 1997. Colour Physics for Industry, Bradford, England, Society of

Dyers and Colorists.

NAPIER, S. 2007. Personal communication, Australia, Biolab Group.

OHNO, Y. 2000. CIE Fundamentals for Color Measurements. IS&T NIP 16

International Conference on Digital Printing Technologies, 540-545.

QIAN, X. Y., WANG, Y. J., WANG, B., F., 2012. Fast color contrast enhancement

method for color night vision. Infrared Physics & Technology, 55, 122-129.

RIGG, B. 2007. Colorimetry and the CIE system. In: MCDONALD, R. (ed.) Colour

Physical for Industry. Bradford, England: Society of Dyers and Colorists.


Springer-Verlag.

SINCLAIR, R. S. 1997. Light, light sources and light interactions. In: MCDONALD,

R. (ed.) Colour Physics for Industry. Second Edition ed. Bradford, England:

Society of Dyers and Colourists.


92

STOCKMAM, A., MACLEOD, D. I. A., JOHNSON, N. E., 1993. Spectral

sensitivities of the human cones. Journal of the Optical Society of America A,

10, 2491-2521.

TASI. 2004. Colour theory: understanding and modelling colour [Online]. Bristol:

University of Bristol. Available:

http://www.tasi.ac.uk/advice/creating/pdf/colour.pdf [Accessed December

5,2007.

VAUGHN, M., VAN OORSCHOT, R., BAINDUR-HUDSON, S., 2008. Hair Color

Measurement and Variation. American Journal of Physical Anthropology 137,

91-96.

VAUGHN, M., VAN OORSCHOT, R., BAINDUR-HUDSON, S., 2009. A

comparasion of hair colour measurement by digital image analysis with

reflective spectrophotometry. Forensic Science International, 183, 97-101.

WOLFRAM, L. J., ALBRECHT, L., 1987. Chemical- and photo-bleaching of brown

and red hair Journal of the Society of Cosmetic Chemists, 38, 179-192.

WOLFRAM, L. J., HALL, K., HUI, I., 1970. The Mechanism of Hair Bleaching.

Journal Cosmetic Chemists, 21, 875-900.

WYSZECKI, G. 2006. Color, Chicago, World Book Inc.

http://www.tasi.ac.uk/advice/creating/pdf/colour.pdf


93

Chapter 5 Thermal Analysis of Human Hair

5.1 Thermal Analysis

Keratin fibres undergo physical and structural changes when they are heated or drawn.

These changes are sometimes accompanied by endothermic or exothermic changes in

the thermal measurement of keratins obtained by differential scanning calorimeter

(DSC) (Konda, 1973).

5.1.1 Differential Scanning Calorimetry (DSC)

Differential scanning calorimetry, according to the ICTA (International Committee for

Thermal Analysis) nomenclature committee, is a technique in which the difference in

heat flux (power) to the sample (pan) and to a reference (pan) is monitored against

time or temperature while the temperature of the sample in a specified atmosphere is

programmed (Groenewoud, 2001). The instrument used is a differential scanning

calorimeter (DSC). The DSC cell is composed of two heating compartments, one is for

the sample and the other for the reference. The reference is usually a sample pan,

which is left empty. There are two principal types of instruments used for DSC

investigations: power-compensation differential scanning calorimetry (power-

compensation DSC) and heat-flux differential scanning calorimetry (heat-flux DSC).

Their schematic experimental setups are shown in the Figure 5.1.

In power-compensation DSC two nearly identical (in terms of heat losses) measuring

holders (microfurnaces) are supplied with individual heaters and their temperatures are

measured with separate sensors. The temperature of both holders can be linearly varied

as a function of time, being controlled by an average-temperature control loop

(Groenewoud, 2001). The detection of any temperature difference between the

furnaces leads to an immediate adjustment of the heating power in such a way that the

temperature is (almost) identical in both furnaces. The additional (or reduced) heating

power is directly correlated with the heat flow into or out of the sample.


94

One single heater is used in the heat-flux DSC to increase the temperature of both

sample cell and reference cell. Small temperature differences, which occur due to

exothermic/endothermic effects in the sample, are recorded as a function of the

programmed temperature (Groenewoud, 2001). This temperature difference is

measured and converted into a heat flow signal through a calibration procedure.

By convention, endothermic events are displayed as upwards signals in power-

compensation DSC curves and as downwards signals in heat-flux DSC curves.

Figure 5.1: Schematic representation of the experimental setup in power compensation

DSC (top) and heat flux DSC (bottom, disk-type system) (Bunjes, 2007).

5.1.2 Thermal Stability of Polymers

Differential Scanning Calorimetry (DSC) is one of the most effective analytical

techniques to characterize the physical properties of polymers. DSC allows

determining heat capacities in both solid and liquid states, phase transition

temperatures and the corresponding enthalpy and entropy changes, as well as heat

capacity changes (Schick, 2002).


95

5.1.2.1 Thermal Stability of Proteins

Proteins as common biopolymers, have complex structures, which include the

numerous levels such as, primary, secondary, tertiary and quaternary structures

(Alberts, 1998). When this structure is destroyed, it may involve partial or entire

unfolding of the protein through breaking the hydrogen bonds, which stabilises the

higher-order native structures of proteins (Figure 5.2). This process is called

denaturation. Denaturation is the process by which a protein is transformed from an

ordered (“native”) state to a less-ordered state due to the rearrangement of hydrogen

bonding without any change to covalent bonds in the polypeptide backbone of the

protein molecule (Bischof, 2002). Since a protein can be either partial or total

denaturated, this process can be reversible or irreversible. In the case of heat

denaturation of proteins at temperature > 45°C, the process is generally considered as

irreversible (Bischof, 2005).

The hydrophobic interaction has a predominant role in determining the stability of the

folded or native state of the protein rather than other interactions, such as hydrogen

bonds, Van der Waals forces and electrostatic interactions. The hydrophobic

interaction between exposed non-polar amino acid residues on the surfaces of the

protein molecule is short-range and orientation dependent (Chiew, 1995). The

maximum stabilities can occur at different temperatures, however, the stability of the

folded state decreases at both higher and lower temperatures, and the denatured

proteins often show irreversible behaviour at high temperature.

In the case of hair proteins, the helical domains of the IFs in the protein try to unfold at

elevated temperatures. The ideal, unfolded protein is the random coil, in which the

rotation angle about each bond of the backbone and side-chains is independent of

bonds distant in the amino acid sequence. In the context of thermodynamics, when

sufficient energy is transferred to a relatively compact ordered protein structure, a

change in the molecular conformation occurs, which results in a more flexible,

disorganized, opened polypeptide chain structure. This is viewed as protein

denaturation. The involved energy usually contains two parts: one is an activation

energy barrier (kinetic), and the other is enthalpic (total heat absorption or release).


96

The kinetic (activation energy) barrier determines the temperature and time

dependence of the denaturation process. As the temperature rises, it becomes

thermodynamically favourable (i.e., sufficient energy to exceed the free-energy barrier

of activation) for the protein to denature (Bischof, 2005). The final denatured state can

be at a higher or lower total energy than the original state. When the final state is at a

lower energy, this involves a strongly exothermic process, such as coagulation,

aggregation, and/or gelation of denatured proteins (Bischof, 2005) (Figure 5.2).

Kinetic models assume that protein denaturation is a rate process. There are many

models for describing the protein thermal denaturation. The simplest, but most widely

used kinetic model to express the ‘‘2-state’’ behaviour is the first order irreversible rate

reaction model. In this case, the process of protein denaturation may be represented by

a transition between two distinguishable states:

UN (5.1)

Where N represents the native or folded state and U represents the unfolded state,

although the compact intermediate state could represent a subset of U, which is

continuously alternating between various energetically unfavourable states (Creighton,

1990).

Figure 5.2: (A) describes the hydrogen bonds of a simple α-helix secondary structure

within one protein. (B) describes the energy states of protein denaturation (not to scale).

There is a state of activation (energy input to the system) followed by denaturation to

an energy state above or below that of the initial natured state (Bischof, 2005).


97

Higher-order kinetic models, which assume that one or more intermediate states exist

between the native and denatured states, were also adopted in several studies

(Lyubarev, 2000, Cravalho, 1992, Sanchez-Ruiz, 1992), which found that higher-order

models could fit experimental results better than the first-order model.

The simple first-order model (Equation.5.1) assumes that each denatured protein

molecule transforms irreversibly in a first order reaction into a species from which the

native form cannot be recovered. A more complex approach is described by the

Lumry-Eyring model (Lumry, 1954). In this model, the irreversible protein

denaturation process involves two steps, (i) a reversible unfolding of the native protein

(N) followed by (ii) a first order, irreversible step, which involves the secondary

structure unfolding to a final state (F), which is unable to fold back to the native one,

as:

(5.2)

Where N, U and F represents native, partially unfolded and final (irreversibly

denatured) protein form; k1, k2 and k are the rate constant for the corresponding

reactions. Comparison shows that the one-step model in Equation 5.1 is a particular

case of the Lumry and Eyring model (Equation.5.2).

There are two main situations when the Lumry and Eyring model (Equation.5.2) is

reduced to a one-step irreversible model (Equation.5.1). In the first case the value of k2

is much higher than the values of k1 and k, so that the direct reaction of the first step is

rate-limiting and the reverse reaction is practically absent. If this step is fast enough,

the DSC transition is entirely determined by the kinetics of formation of the final state

and equilibrium thermodynamics analysis is not permissible. Another situation

happens when the rates of the direct and reverse reactions of the first step (k1 and k2)

are much higher than the rate of the second step (k), but equilibrium for the first step is

shifted toward the form N (Sanchez-Ruiz, 1992).


98

5.1.2.1.1 Thermal Stability of Keratin Fibres

DSC studies on a wide variety of keratin materials have been conducted by

investigating the thermal properties of the main morphological components, namely of

IFs and matrix (IFAP), in wool and other keratinous fibres. Schwenker et al (1960)

verified that, when a keratin fibre is heated it goes through a number of changes/phases

before it eventual degrades to charred residue.

The thermal analysis by DSC allows investigation of the structural changes and the

degree of degradation of hair keratin based on the enthalpy data and denaturation

temperature of keratin. Assuming the validity of Feughelman’s (1959) two-phase

model (intermediate filament and matrix) and a 2-state transition model (native and

denatured) for the denaturation transition in the keratin fibres, the denaturation

enthalpy reflects the progress of the helix-coil transition in the crystalline sections of

the intermediate filaments (Wortmann, 1993). In the IFs the concentration of cystine is

low and the location and distribution of the sulphur crosslinks are known to be material

invariant (Sparrow, 1988). On the other hand, the cystine crosslinks are mainly present

outside the helical regions in the matrix material. A significant positive and largely

linear correlation was found between denaturation temperature and cystine content for

various keratins (Spei, 1989, Crighton, 1985), so that the denaturation temperatures for

the various keratins can largely be attributed to the matrix material and mainly to its

cystine and thus crosslink concentration. In this way, the denaturation enthalpy

depends on the stability of the intermediate filaments, which are composed of

α−helical proteins, while denaturation temperature is kinetically controlled by the

crosslink density of the matrix, in which the intermediate filaments are embedded

(Wortmann, 2002). The limiting step of the thermal denaturation process is considered

to be the fission of S–S bonds between the IF and matrix (IFAP) (Istrate, 2009).

The theory that the denaturation enthalpy is related to the amount and structural

integrity of α-helical material in the IFs while the denaturation temperature reflects the

thermal stability of the matrix has been applied to interpret the effects of cosmetic

treatments. Leroy et al. (1992) observed a decrease in denaturation peak area through

bleaching. This can be attributed to a loss of crystalline material, as observed by SAXS

(small angle X-ray scattering) (Franbourg, 1996) or to a general decrease in native α-


99

helical material, which can be denatured (Schmidt, 1994). Wortmann et al. (2002)

realized that after excessive bleaching and permanent wave treatments, the

denaturation temperature and enthalpies were reduced for both types of samples. It was

also concluded by Popescu et al. that bleaching affects the amorphous matrix and

α−helix faster than the permanent wave treatment (Popescu, 2003).

However extensive damage to the α-helical material is not necessarily accompanied by

a loss of X-ray crystallinity (Kanetaka, 1993). Kuzuhara (2006, 2007) used raman

spectroscopy on human hair to measure the protein denaturation from α- to random

form by oxidation and reduction, to prove the “native helix” hypothesis, which was

originally proposed by Wortmann (2002, 2006). They believed that the denaturation

enthalpy relates in fact only to the amount of “native” α-helical material, which still

shows a denaturation transition, regardless of the total amount of α-helical material in

hair, which may already be chemically degraded to some extent.

5.1.3 DSC Investigation of Keratins

Different techniques of thermal analysis of keratin reveal different endothermic

transitions. Some environmental effects, which affect the endotherms of keratin fibres

in the DSC measurement, should be taken into account, such as DSC cell types,

sample preparation, heating rates, the use of open or sealed vessels, and thermal media.

Water and oxygen are two of the factors that influence the degradation of keratin by

heat, thus the thermal behaviour of keratin fibres were studied by two DSC methods,

“dry DSC” and “wet DSC”. The effect of damage is evaluated while the keratin fibre is

heated with the use of an open or closed system.

5.1.3.1 DSC Investigations of Keratin in the Dry State

The “dry DSC” method allows the moisture content of the keratin sample to evaporate

with increasing temperature. Endothermal effects sometimes appear as a doublet above

200oC (Haly, 1967, Spei, 1989). Spei and Holzem (1987) have shown that the


100

denaturation peak can usually be detected adequately and evaluated for dry fibres, but

that the effect is always secondary in size compared to a large background peak due to

general keratin pyrolysis (Wortmann, 2002). Leroy et al (1992) investigated virgin,

bleached and perm-waved hair by DSC in the dry state. They observed that with

bleaching the DSC peak for dry fibres shifts to higher temperature and the denaturation

peak area decreases. Cao (1999) developed a method, using silicone oil as the thermal

medium for human hair in DSC analysis, to avoid the water evaporation.

5.1.3.2 DSC Investigation of Keratin in the Wet State

Keratin is known to become softer with water uptake (decrease in modulus) (Milczarek,

1992). Most thermal transitions analysis have been conducted on samples placed under

vacuum, immersed in water, or enclosed in a container with a known amount of H2O.

In this closed system the lack of oxygen may avoid the oxidative reaction, although the

interaction with water and other pyrolysis products may accelerate the decomposition.

The properties of hair, which has been completely dried on vacuum or highly

plasticized by immersion in water under pressure, are drastically different compared to

ambient conditions. At ambient conditions keratin contains considerable amounts of

water, which form an integral part of its structure. It has been indicated that thermal

transitions in keratin are strongly affected by water and depend on the amount present

(Milczarek, 1992).

It is generally assumed that water absorbed in keratin exists in three forms: a) water

adsorbed on strongly binding sites; b) water adsorbed on weakly binding sites and/or

hydrogen bound to strongly adsorbed water; c) loosely bound or free water (Sakabe,

1987, Watt, 1980). At temperatures around 100oC a broad endothermic DSC maximum

signal is often observed (Schwenker, 1960, Humphries, 1972), which is related to the

removal of free and weakly bound water. The liberation of strongly bound water

happens at temperatures above 140 oC (Watt, 1959). Horio et al. (1965) found that the

loss of cystine caused by heating at 140 oC was greatly increased by an increase in the

amount of water sealed with the sample.


101

In soluble proteins the helix thermally denatures predominantly at temperatures below

80oC, but the insoluble keratin proteins denature above 100

oC. This is ascribed to the

rigidity of the matrix, whose viscosity impedes the segmental mobility of the α-helix

and the unfolding process (denaturation) (Istrate, 2009). Previous studies demonstrated

that the endothermic peak was observed at 230-250°C when the keratin fibre was dry

but appeared at 120-150°C, when the fibres were wet (Haly, 1967, Crighton, 1990).

This stabilization of the helix at high temperatures is attributed to the cross-link

density of the matrix, into which the IFs are embedded, thus significantly enhancing its

temperature tolerance. Since water acts an effective plasticizer in the matrix

(Wortmann, 1985, Feughelman, 1997), the denaturation temperature decreases

accordingly with increasing water content (Haly, 1967), while the denaturation

enthalpy remains constant (Cao, 2005).

Milczareck et al. (1992) verified that water stabilizes the crystalline structure of keratin,

while the partial removal of water disorganizes the crystalline phase. Hair samples kept

in increasing relative humidities exhibited an increase in the enthalpy of water

evaporation.

5.1.3.3 Two Hypothesis in the Peak Separation

Since Haly and Snaith (1967) first reported the observation of a bimodal endotherm in

DSC studies of wool, this endothermic doublet has been observed in the range of the

degradation temperature of the α-helix in cortical cells, and interpreted using two

principally different theories.

Spei and his coworkers conducted DSC studies in the dry state on the various keratins

(1990), as well as on isolated low-sulfur and high-sulfur proteins from Lincoln wool

(1983, 1989). They verified that in case of endothermic doublets, the low-temperature

peak around 230-240 oC is attributed to helix denaturation, while the high temperature

peak (≈250-260oC) corresponds to the matrix degradation from cystine pyrolysis

(1987).

Cao et al (1997a, 1997b) examined Merino wool in an intermediate state

between dry and wet in silicon oil, which is used to preserve a certain amount of water

in the fibres. They found that there is a shift of the endotherms to lower temperatures


102

compared to the dry state. They considered the lower temperature peak of the doublet

at ≈ 170oC (dT/dt = 5

oC/min) as originating from melting or rather denaturation of the

α-helical, crystalline material of wool keratin. The broad, higher temperature

endotherm beyond around 185oC they considered as the thermal degradation of other

histological components, which is in a good agreement with Spei’s finding. This

endotherm doublet interpretation is referred to as the “helix/matrix” hypothesis.

The helix/matrix hypothesis attributes a second peak or a shoulder to the matrix

degradation. However, this might due to the denaturation of the para-cortex,

introduced by the ortho/para-cortical hypothesis. Taking Cao’s (1997a) finding as an

example, the shape of a bimodal curve is explained by three peaks; ortho-cortex, para-

cortex and their overlap, with a relatively small cortex denaturation peak.

Wortmann and Deutz (1998) confirmed the ortho/para-cortical hypothesis, which was

proposed initially by Haly and Snaith (1967), Crighton and Hole (1985) and Crighton

(1990). They suggested that the appearance of double denaturation endotherms for

wool is due to the differences of the cystine content and disulfide linkages in the

matrix of ortho- and para-cortical cell, which are pronounced enough to allow the

separation of the peaks. Isolated ortho-cortical cells in water show a denaturation

temperature at about 138°C, whereas the para-cortical cells have a TD at about 144°C.

The higher temperature at which the para-cells denature was attributed to a higher

concentration of disulfide linkages (or cystine) (Wortmann, 1998). These values were

in good agreement with the results obtained from the whole wool structure. The

formation of two separate peaks has also been observed in the dry state, where the

denaturation peaks yielded 230°C and 240°C. This again relates to the ortho- and para-

cortical cells, respectively (Wortmann, 1998).

Tonin et al. (2004) confirmed the existence of the endotherm double peak for wool

keratin and verified the decrease of denaturation in the first peak referring to the ortho

cells without changes in the second endotherm peak for treated wool. However, other

studies show that the bimodal endotherm of wool in water excess depends on the

environment and the heating rate of DSC measurements (Cao, 1997a).


103

5.1.4 Nonisothermal Kinetics of Keratin Thermal Denaturation

Every chemical reaction is associated with a certain heat of reaction, consequently the

heat flow rate is proportional to the rate of reaction. The assignment of the time

dependence heat flow rates to defined reactions leads to kinetic data. The advantage of

DSC in solving the kinetic problems is its simple, fast sample preparation and the wide

range of experimental conditions, which can produce the information in a short time. It

will immediately give a series of “reaction rates” as a function of the extent of reaction,

depending on time and temperature. The aim of kinetic investigations is then to find

quantitatively this functional relationship.

DSCs are generally operated by a controlled program which changes the temperature

with time. There are two distinguished modes of operation, one is the constant heating

rate (i.e., the classical DSC operation mode) and one is the variable heating rate

(periodical or non-periodical change).

5.1.4.1 Constant Heating Rate

In this mode of operation the controlled program follows the time law:

T = T0 + βt (5.3)

Where, T = temperature

T0 = starting temperature

β = the heating rate, dT/dt,

t = time

The temperature thus changes linearly in time. Because heat capacity can be defined as

the amount of energy required to increase the temperature of a material by 1 Kelvin or

degree Celsius, thus:

Cp = Q/∆T (5.4)

Where, Cp = the heat capacity


104

∆T = the change in temperature

Q = amount of heat required to achieve ∆T

When temperature is changing, the rate of heat flow is given by

dQ/dt = Cp dT/dt (5.5)

With the heating rate β = dT/dt , Equation 5.5 leads to

dQ/dt = β Cp (5.6)

or

Cp = (dQ/dt)/β

This provides one way of measuring heat capacity in a linear rising temperature

experiment: one simply divides the heat flow by the heating rate. If the temperature

programme is replaced by one comprising a linear temperature ramp modulated by a

sine wave, this can be expressed as:

T = T0 + βt + B sin ωt (5.7)

Where, B = the amplitude of the modulation

ω = the angular frequency of the modulation

And this is the simplest and most used modulation type, namely, a periodic (harmonic,

i.e., sinusoidal) temperature programme. In addition, B and ω together with β as the

“underlying” heating rate, can be chosen freely within certain limits. The derivative of

Equation.5.7 with respect to time is

dT/dt = β + ω Bcos ωt (5.8)

Thus, it follows with Equation.5.5:

dQ/dt = Cp(β + ω Bcos ωt) (5.9)


105

5. 1.4.2 Variable Heating Rate (Modulated Temperature)

In a conventional DSC, a constant heating rate is applied. Although there are many

different forms of temperature programme, a sinusoidal temperature modulation is the

most popular way, as illustrated in Figure 5.3. Modulated DSC, which uses a

sinusoidally modulated temperature profile, superimposed a sine wave onto a

conventional constant heating rate experiment.

Figure 5.3: Temperature-modulated DSC heating profile (Gill, 1993)

The raw data are averaged over the period of one oscillation to remove the effect of

modulation and provide the output dQ/dt, which is referred as the average or total heat

flow. This output is equivalent to that of conventional DSC and yields the total heat

capacity CpT as a function of temperature through:

CpT

=1/(β m) dQ/dt (5.10)

Where, β = dT/dt is the heating rate

t = time

T = absolute temperature

m = sample mass


106

This curve for total heat flow is subtracted from the raw data and the modulation is then

subjected to Fourier Transform Analysis to obtain the amplitude and phase difference

of the heat flow response with respect to the temperature modulation. Through

consideration of the cyclic heat capacity or modulus of complex heat capacity, it finally

yields the so-called reversing heat capacity of the sample CpR (Gill, 1993, Schawe,

1996).

If the results are expressed as heat capacities, then the average total heat flow is divided

by the underlying heating rate β. Thus,

(∆dQ/dt)/β = CpT = the average or total heat capacity

Furthermore, CpT

is simplified (Schawe, 1996) by two components as:

CpT = Cp

R + Cp

NR (5.11)

where CpNR

is called as non-reversing (NR) or less commonly the kinetic heat capacity.

Although all of the parameters in Equation. 5.11 are expressed as heat capacities, they

can equally well be expressed as heat flows.

dQ/dt = average or total heat flow

CpR β = reversing heat flow

CpNR

β = non-reversing heat flow

If the underlying heat flow is changing slowly and smoothly under the modulation,

averaging the modulated signal over the period of the modulation will approximately

provides the same information as an un-modulated experiment. And the ”averaging”

means the modulated experiment looks ‘smoothed’ to some extent.


107

5.1.4.3 Kinetic Investigation

While the sample is subjected to a controlled temperature ramp, it is reasonable to

assume that it reflects a transformation which can be represented for the simplest case

as:

B0 B1 (5.12)

Where B0 is the material before transformation and B1 is the material after

transformation. Any conversion is accompanied by an absorption or release of heat. In

a quantitative way it is expressed by means of the enthalpy of the process (Wortmann,

2006):

dtdTH

t

t

p

T

T

p CC

2

1

2

1

(5.13)

Where, ∆H = enthalpy

t = time

T = temperature

Cp = heat capacity

β is the heating rate (Equation.5.3), thus, the degree of conversion, α, is calculated

from the DSC curve assuming that the area under the peak up to a given time is

proportional to the degree of conversion:

total

t

PA

PA (5.14)

Where PAt is the integral of the peak up to time t and PAtotal is the overall peak area.

With this definition of conversion, the reaction rate is given by the following equation

(Istrate, 2009):

α

= M (α (T)) = k (T) f (α) (5.15)


108

Where α

is the reaction rate, α is the degree of conversion, k (T) is the rate constant as

a function of the absolute temperature T, and f(α) is an unknown function of

conversion.

The temperature dependence of the rate constant k (T) is usually described by the

empirical Arrhenius equation or less frequently by the Eyring equation, which follows

from the theory of the activated complex:

Arrhenius: k (T) = A∙ exp )(RT

AE (5.16)

Where, A = preexponential factor

EA = activation energy

R = universal gas constant

Eyring: k (T) = )(h

TKB exp

) ∙ exp (─

) (5.17)

Where, KB = Boltzmann constant

h = Planck constant

∆S = activation entropy

∆H = activation enthalpy

Here, ∆H =EA ─ R∙T

∆S= R ∙ ln )(TKe

hA

B

The inseparable combination of rate law f(α), activation energy EA, and preexponential

factor A is sometimes referred to as the “kinetic triplet” (Maciejewski, 2000).

There are several methods for calculating the kinetic triplet, f(a), EA and A. A certain

reaction model implies a specific analytical form of the kinetic function f(a). There are

also the model-free methods which allow calculating the value of the activation energy,

EA, without any knowledge of the reaction pathway. The advantage of analysing

kinetic data using model-free methods is their independence of any model or

mechanism, and they are thus able to describe the most complicated reaction behaviour


109

at different temperatures (Fernandez d’Arlas, 2007, Vyazovkin, 2006), as long as the

reaction mechanism does not change with temperature and heating rate (Fernandez

d’Arlas, 2007).

5.1.4.4 Kinetic Analysis Methods

It is necessary to discuss the kinetic analysis methods in relation to the temperature

control modes. From the early 1970’s, ASTM International has developed a series of

kinetic methods to meet a variety of needs, these include E2041 which makes use of a

single scan using the Borchardt and Daniels approach (1957), E2070 using the

isothermal Šestak-Berggren approach (1971), and E698 using the multiple heating

rates Ozawa (1970) and Flynn and Wall (1966b) approach.

The isoconversional method of Flynn, Wall and Ozawa (Flynn, 1966a, Ozawa, 1970)

is a ‘model-free’ method, which involves measuring the temperatures corresponding to

fixed values of α from experiments at different heating rates β and plotting ln(α)

against 1/T. The slopes of such plots give -EA/R. If EA varies with α, the results should

be interpreted in terms of multi-step reaction mechanisms (Burnham, 1999, Ozawa,

1970).

ASTM E698, as reviewed by Popescu and Segal (1998), is based on the Arrhenius-

equation, which is used to investigate the dependence of the denaturation temperature

TD on the heating rate. The Arrhenius-equation for this interrelation of these two

parameters can be expressed as followed:

β = A exp [-EA/ (RTD)] (5.18)

Where A is the pre-exponential factor, EA is the activation energy of the process

(J/mol), and R is the molar gas constant (8.314J/Kmol). EA is determined from the

slope of the line through the data in a so-called Arrhenius-plot, which is represented

as ln β vs. 1/TD.


110

5.2 Objectives

DSC is the technique of choice to investigate the calorimetric effects associated with

protein structural changes (Bischof, 2005). In this chapter thermal analysis was

performed on bleached hair in water to determine the structural changes and the degree

of degradation based on the enthalpy and denaturation temperature data and to discuss

them in terms of the thermal properties of filaments and matrix. A kinetic approach

based on ASTM E698 for the description of the oxidation influences is applied.

5.3 Experimental

There are generally two ways of measuring DSC of keratins, namely in water and in a

dry environment (allowing the moisture to evaporate during heating). The DSC under

dry condition was shown to give misleading information due to the interference of

pyrolysis (Wortmann, 2002). Consequently, the present work deals exclusively with

DSC of keratins in water.

There are two types of experiments performed. In the standard Modulated DSC mode

(MDSC), the samples (6% H2O2 bleached hair, 9% H2O2 commercial bleached hair and

commercial persulphate bleached hair) were heated with the temperature increasing

linearly at a rate of 3oC/min, and the corresponding heat flow was recorded. In the

kinetic investigation, the samples (virgin human hair, mohair and 2h commercial

persulphate bleached human hair) were heated with various heating rates between

0.1oC and 7

oC, applying modulation parameters suitable for a given heating rate, as

shown in the Table 5.1.


111

Table 5.1: The parameters for the heating rate and modulation

Heating Rate(oC/min) 0.15 0.2 0.3 0.5 1 1.5

Modulation (oC/min) 0.05 0.1 0.1 0.1 0.5 0.5

Heating Rate(oC/min) 2 2.5 3 4 5 7

Modulation (oC/min) 1 1 1 1 1 3

All analyses were conducted on a power-compensated DSC instrument (DSC Q10,

TA-Instruments), using pressure-resistant (25bar), stainless-steel, large-volume

capsules (TA Instruments) in the temperature range of 80-180oC in excess water (high

pressure DSC). Because the initial evaporation of water quickly builds up high

pressure inside the pan, it suppresses further evaporation.

Randomly selected hairs taken from the midsection of a hair tress were used to

produce hair snippets (~ 2mm in length). Prior to measurement the samples were

stored under standard room condition (20oC, 65%RH) to ensure invariant water

content. Hair snippet samples between 3 and 10 mg were weighted into the sample

container. The weight basis for all calculations was the conditioned weight of hair,

containing 12% water (Wortmann, 2007). 40 μl water were added, the container was

sealed with a lid using the sample encapsulating press, and the samples were stored

overnight to achieve equilibrium water content and distribution. As a reference, an

empty container was used. At least two samples for each experiment were measured.

Two assumptions have to be considered in determining the kinetics by DSC: (1) the

heat flow relative to the instrument baseline is proportional to reaction rate and (2) the

temperature gradients through the sample and the sample-reference temperature

difference is small. By using small sample sizes (<10mg) and slow heating rates (or

isothermal experiments) these assumptions are reasonable.

The collected DSC raw data were analysed using the TA Instruments Universal

Analysis 2000 Program. The curves were processed using suitable, nonlinear baselines

provided by software PeakFit V4.12, which led to improved estimates for the

characteristic parameters. Processing the modulated DSC-data by the instrument’s


112

software yielded curves for total heat capacity, (CpT), reversing heat capacity, (Cp

R)

and non-reversing heat capacity, (CpNR

). From the CpR

curve the heat capacity change

∆Cp, which is underlying the phase transition, is provided. The curve of CpNR

yields the

peak temperature TD and the peak area with the determined baseline, which is referred

to as the denaturation enthalpy ∆HD. Further data analyses were conducted using Excel

software (Microsoft).

5.4 Results and Discussion

5.4.1 DSC of Virgin Hair in Water

By applying DSC to human hair in water, the thermal stabilities of the major

morphological components in the cortex can be determined. All DSC curves recorded

at 3oC/min heating rate for the hair samples exhibit the endothermic process. An

example of untreated hair, for which the endothermal effect is observed at around

144°C, is illustrated in Figure 5.4.

The α-keratin denaturation peak of human hair is not reversible. When an endothermic

peak becomes apparent in the curve of the non-reversing heat capacity CpNR

, the

denaturation enthalpy ΔHD is determined by applying a linear baseline between the

start (T1) and end temperature (T2) of the peak. The initial data obtained from the

DSC-curves are hence the peak- or denaturation temperature TD and the denaturation

enthalpy ΔHD. The start and end temperature of the baseline, defining the peak, are

marked in Figure 5.4.


113


DSC curve for untreated human hair in water.

The green curve represents the overall heat capacity of the human hair whereas the

blue curve is the non-reversible heat capacity of the human hair.

The denaturation temperature and the peak area reflect different aspects of the thermal

stability of the matrix and the filaments in the cortex of hair, respectively. The peak

temperature, TD, which is the denaturation temperature of the helical material, is

influenced by the heating rate (Rüegg, 1977) as well as the surrounding environment

(especially pH) (Bischof, 2005, Eyring, 1974, Joly, 1965). The denaturation enthalpy

∆HD, which is the energy required for the helix denaturation, is independent of any

denaturation model assumption (Cooper, 1999). It was shown by Spei et al (1987) that

the area of the peak related to the keratin denaturation is a good measurement of the α-

helix content. The recorded endotherm reflects the progress of the helix-coil transition

in the crystalline sections of the intermediate filaments. However the endotherm varies

in terms of shape and size from measurement to measurement (Cao, 1999).

There are several methods of determining the area of a DSC peak, depending on the

selected shape of the baseline. The standard method is to use a straight baseline, while

T2

T1


114

in some cases a parabolic or hyperbolic baseline is applied. Milczarek et al (1992)

found that the use of a hyperbolic shape of the baseline causes only a ~5% differences

in the obtained peak area, compared with the linear baseline. Thus, the standard linear

baseline is adapted in our thermal analysis study, which is provided by the TA

Instruments Universal Analysis 2000 Program.

The reversible heat capacity of virgin hair is given by the MDSC-curve in the wet state

(Figure 5.5). This CpR curve shows that there is a stepwise change in the range of the

heat capacity of 1.14 J/(g∙K). The ∆Cp R

is decided by measuring the distance between

the upper and lower baselines, where the horizontal lines (- - -) define the anticipated

maximum point and lowest point in the transition change. The non-reversible heat

capacity ∆CpNR

also is shown in the Figure 5.5. It is obvious that the two points which

define the peak area on the non-reversible heat capacity curve correspond to similar

temperatures with the maximum and lowest point in the phase transition change on the

curve of reversible heat capacity in human hair.

Figure 5.5: Typical reversing CpR and non-reversing Cp

NR DSC curves for untreated

human hair in water. The green curve represents the curve of reversible heat capacity

for human hair, whereas the blue curve shows the non-reversible heat capacity. The

upper and lower baselines for the step-wise change are indicated and the value of ΔCpR

is given.

∆CpR

=1.14 J/(g∙oC)


115

In the further discussion, we will only concentrate on analyzing the thermal reaction in

the DSC peak occurring in the 100-180oC temperature range (α-keratin or crystalline

peak).

5.4.2 Water Absorption Effect on the Keratin

In the process of sample preparation (Chapter 2), 9% H2O2 commercial bleached

sample (3.5h, 4h) and commercial persulphate bleached sample (3.5h, 4h) were left in

water overnight to investigate the influence of thermal analysis of soaking keratin in

water for prolonged times.

The commercial persulphate bleached hairs were chosen as an example. The samples

which had been bleached previously for 2.5h, 3h, 3.5h, and 4h were stored in water for

an additional 48 hours at room temperature. The denaturation temperatures TD and

denaturation enthalpies ∆HD of keratin were correspondingly compared, and the results

are presented in the Table 5.2.

Table 5.2: The comparison between denaturation temperatures TD and enthalpies ∆HD

for commercial persulphate bleached hairs before and after soaking in water for 48

hours

Bleaching

Time (h)

TD (oC) ∆HD (J/g)

After

Bleaching

Soak in Water

(48h)

After

Bleaching

Soak in Water

(48h)

2.5 125.1±0.1 127.6±2.0 9.8±0.3 10.4±0.0

3 124.7±0.1 129.3±1.2 10.9±0.0 10.9±0.4

3.5 127.3±1.7 130.9±4.1 9.2±0.4 9.8±0.3

4 129.9±0.7 130.3±0.5 11.2±0.7 10.3±0.4

Although part of the α-helical phase is not accessible to water at room temperature

(Feughelman, 1989), the individual microfibrils still show some degree of swelling

(Spei, 1984), thus water may still access the α-helical regions of the microfibrils. This

swelling is considerably less than that of the matrix, in which they are embedded. As


116

described in the previous literature discussion, the denaturation temperature of the

helices is controlled by the viscosity of the surrounding matrix. The denaturation

temperature drops accordingly with the water content, when water acts as a plasticiser

in the matrix. This finding was confirmed by Haly and Snaith (1967) by DTA

experiments on wool. Similar experiments also were conducted by Cao and Leroy

(2005) on human hair by DSC. They believe that the denaturation enthalpy is not

affected by a change of water content. This conclusion is reflected in the results in

Table 5.3, which were obtained before and after soaking keratin in water.

In Table 5.3, all of the bleached samples follow the same trend, that there is a

consistent increase in TD after soaking in water. Little effect on ∆HD is observed, giving

the evidence that water plays a very important role for the denaturation temperature,

which is kinetically controlled by crosslink density of the matrix (Wortmann, 2002).

Due to the above reason, the results of denaturation temperature and denaturation

enthalpy at 3.5h and 4h were excluded from further thermal transition investigation, to

avoid the unexpected results, since the storage time in water strongly affects the

thermal transitions of keratin (Milczarek, 1992).

5.4.3 DSC on Various Bleaches Effects

DSC measures the effects of bleaching on the DSC curves of hair fibres in water. The

effects, as reflected in changes of denaturation temperatures TD and enthalpies ∆HD,

are discussed in terms of the thermal properties of filaments and matrix.

There are two hair sources involved in the bleaching procedure (see Chapter 2).

Because of the differences between hair batches, the relative denaturation temperature

(TDR)

and the relative denaturation enthalpy (∆HDR) were employed in this

investigation, given by

TDR

= (TD/TD

0) (5.20)

Where, TD0

is the denaturation temperature of virgin hair


117

TD is the denaturation temperature of target hair

And,

∆HDR

= (∆HD/∆HD0) (5.21)

Where, ∆HD0 is the denaturation enthalpy of virgin hair

∆HD is the denaturation enthalpy of target hair

In addition, the results for bleached hair samples which had been treated for 3.5h and

4h were taken off, because of the water storage effect (see section 5.4.2).

Tables 5.3-5.5 summarize the DSC results of denaturation temperature and

denaturation enthalpy for the bleached samples for the various bleaching treatments.

The values of TDR

and ∆HDR

show a consistent decrease for all bleached hair samples.

For a given bleaching time, commercial persulphate bleached hair shows a more

pronounced damage effect than 9% H2O2 commercial bleach and 6% H2O2 bleached

hair. The aggressive chemical treatment thus results in a lower enthalpy and a lower

peak temperature, which implies less crystalline material in the IFs and less cross-link

density in the matrix (IFAPs). In parallel, the longer the treatment time, the greater is

the structural damage in the hair keratin. Commercial persulphate bleached hair (3h)

has the lowest denaturation temperatures TD value (123.9oC) compared to all other

bleached hairs.

Table 5.3: The results for TD and ∆HD for the 6% H2O2 bleached samples, in the form

of the arithmetic means, as a representative of TD R and ∆HD

R

Bleaching Time (h) TD (oC) TD

R ∆HD (J/g) ∆HD

R

0 145.4±0.4 1 11.5±0.1 1

0.5 142.1±0.7 0.98 10.9±0.3 0.95

1 140.8±0.3 0.97 11.1±0.5 0.97

1.5 141.9±0.7 0.98 10.6±0.1 0.92

2 140.3±0.2 0.97 11.2±0.1 0.98

2.5 140.5±0.5 0.97 11.0±0.7 0.96

3 140.0±0.3 0.96 11.1±0.6 0.96


118

Table 5.4: The results for TD and ∆HD for the 9% commercial bleached samples, in the

form of the arithmetic means, as a representative of TD R and ∆HD

R


R ∆HD (J/g) ∆HD

R

0 145.4±0.4 1 11.5±0.1 1

0.5 141.2±0.4 0.97 10.7±0.1 0.94

1 139.2±0.4 0.96 10.4±0.3 0.91

1.5 137.6±0.0 0.95 11.4±0.2 0.99

2 134.3±0.2 0.92 9.8±0.2 0.85

2.5 132.1±0.1 0.91 4.4±1.2 0.39

3 132.7±0.6 0.91 5.8±0.5 0.5

Table 5.5: The results for TD and ∆HD for the commercial persulphate bleached

samples, in the form of the arithmetic means, as a representative of TD R and ∆HD

R


R ∆HD (J/g) ∆HD

R

0 144.2±0.3 1 15.4±0.5 1

0.5 136.7±0.2 0.95 13.8±0.6 0.9

1 128.8±0.5 0.89 11.9±0.2 0.77

1.5 126.6±0.1 0.88 12.0±0.9 0.78

2 125.0±0.4 0.87 9.6±0.2 0.62

2.5 125.1±0.1 0.87 9.8±0.3 0.64

3 124.7±0.1 0.86 10.9±0.0 0.71

While comparing the three bleaching treatments (6% H2O2, 9% H2O2 commercial

bleach and commercial persulphate bleach), a consistent decrease in TDR as well as

∆HDR with prolonged bleaching time is observed in Figures 5.6 and 5.7, respectively.

Commercial persulphate bleach decreases fastest in both TDR and ∆HD

R than 9% H2O2

commercial bleach and 6% H2O2 bleach, although ∆HDR of 9% H2O2 commercial

bleach (Figure 5.7) drops faster after 2h than commercial persulphate bleach,

indicating that the IF of 9% H2O2 commercial bleached hair is heavily damaged after

the extensive oxidation time (2h). Commercial persulphate bleach and 6% H2O2

arrived their saturation within the 3 hours treatment, however 9% H2O2 commercial

decrease linearly with the increasing time.


119

The curve of 6% H202 bleach illustrates a quite stable value for TD R and ∆HD

R in

Figures 5.6 and 5.7, implying that it has little damage effect on the keratin, or in

another way that it plays little role for affecting both TD R and

∆HD

R.

Figure 5.6: Relative denaturation temperature TDR for 3 bleaches: (6% H2O2 bleach, 9%

H2O2 commercial bleach, commercial persulphate bleach) for 3 hours bleaching time.

The trendline curves are empirical and based on polynomial relationships.

Figure 5.7: Relative denaturation enthalpy ∆HDR for 3 bleaches: (6% H2O2 bleach, 9%

H2O2 commercial bleach, commercial persulphate bleach) for 3 hours bleaching time.

The trendline curves are empirical and based on polynomial relationships.

0.8

0.85

0.9

0.95

1

0 1 2 3

T DR

Bleaching Time (h)

6% H202 Bleach



0

0.2

0.4

0.6

0.8

1

1.2

0 0.5 1 1.5 2 2.5 3

∆H

DR

Bleaching Time (h)

6% H202 Bleach




120

DSC analyses confirm the damage of bleaching to hair keratin. The results show a

lower keratin denaturation enthalpy, which is due to the increase in the disorganization

of IF in the keratin structure, and a lower denaturation temperature, which is related to

the reorganization of matrix crosslinks after the denaturation (Monteiro, 2005). The

lower ΔHD value, the less amounts of the α-form crystallites remain in the IFs.

Likewise, the lower TD value, the lower cross-link density in the matrix (IFAPs)

(Wortmann, 2002). This phenomena is well explained by that bleaching leads to the

increased concentrations of cysteic acid and thus to more ionic interactions, but this

interaction will be broken in the wet state, the increase in the content of anionic groups

induces an increase in water content, leading to a continuous decrease of matrix

viscosity and thus of TD (Wortmann, 2002).

The correlation between ∆HDR

and TDR for the three bleaching treatments is explored in

Figure 5.8 by plotting ∆HDR

against TDR. A strong linear relationship is observed for 6%

H2O2 bleach, commercial persulphate bleach and part of 9% H2O2 commercial bleach,

suggesting that the three bleaches have a homogeneous oxidation effect on IFs and

IFAPs. However the results for 9% H2O2 commercial bleach after 2h bleaching time

are unexpected and it is considered that IFs are subjected to progressive damage after

excessive bleaching treatment. Thus, there is a bigger decrease in denaturation

enthalpy compare with the denaturation temperature. Consequently, the gradient of the

denaturation enthalpy vs. denaturation temperature is prone to be lower.

The results of 6% H202 bleach cluster in a narrow range of the ∆HDR

and TDR

in Figure

5.8, while both 9% H2O2 commercial bleach and commercial persulphate bleach

distribute their denaturation values along both axes, except for the deviative results for

9% H2O2 commercial bleach. This conclusion further confirms that 6% H202 bleach

only has a moderate damage effect on the keratin, whereas there is a prominent

structural change on both IF and matrix by commercial persulphate bleach and 9%

H2O2 commercial bleach.


121

Figure 5.8: Plot of ∆HDR

against TDR for the 3 bleaches: (6% H2O2 bleach, 9% H202

commercial bleach, commercial persulphate bleach) for up to 3 hours bleaching time.

5.4.4 Nonisothermal Kinetics of α-keratin Thermal Denaturation

5.4.4.1 Nonisothermal Kinetics on the Bleaching Effect in Human Hair

The course of the denaturation was investigated by a kinetic analysis of the DSC-

curves for untreated and oxidised hair. There is a consensus that single-heating rate

scans have severe limitations for a kinetic analysis (Opfermann, 2000, Burnham, 2000).

The kinetic analysis of the endothermic process in the DSC measurements is thus

carried out for different heating rates. Ten heating rates were utilized and the results

for virgin hair and commercially persulphate bleached (2h) hair are presented in Tables

5.6 and 5.7.

0

0.2

0.4

0.6

0.8

1

1.2

0.8 0.85 0.9 0.95 1 1.05

∆H

DR

TDR

6% H202 Bleach




122

Table 5.6: Results for virgin hair at various heating rates

Heating Rate(oC/min) 0.15 0.5 1 1.5 2 2.5 3 4 5 7

Modulation (oC/min) 0.05 0.1 0.5 0.5 1 1 1 1 1 3

∆T (K)

128.4

128.3

133.1

133.8

138.1

140.6

137.8

138.4

138.4

140.4

139.8

141.0

141.0

143.6

146.4

144.2

142.9

145.4

145.6

147.0

148.5

147.8

∆H(J/mol)

5.0

6.1

9.3

8.6

11.3

12.2

11.0

15.8

14.2

15.5

15.6

15.4

13.4

13.4

11.7

12.7

11.8

12.4

10.5

11.5

11.9

14.4

Table 5.7: Results for commercial bleached human hair (2h) at various heating rates

Heating Rate

(oC/min)

0.15 0.2 0.3 0.5 1 1.5 2 3 4 5 7

Modulation

(oC/min)

0.05 0.1 0.1 0.1 0.5 0.5 1 1 1 1 3

∆T (K)

113.5

113.4

113.3

114.1

114.4

113.7

116.1

115.6

116.2

117.5

117.4

117.7

120.8

120.5

120.5

122.6

122.7

122.4

124.1

123.7

123.8

126.2

126.4

126.6

127.5

126.7

127.4

129.0

127.6

128.7

129.9

129.3

129.9

∆H(J/mol)

1.9

2.2

2.2

4.0

3.1

3.7

3.7

4.3

4.1

5.8

6.7

6.6

6.9

7.8

6.8

7.4

7.0

6.3

6.5

6.3

7.2

7.9

7.9

7.8

7.7

8.5

8.5

7.8

7.6

8.9

9.7

9.9

7.5


123

The DSC curves used to create Tables 5.6 and 5.7 were analysed on the basis of

ASTM-E698. The DSC curves recorded at various heating rates for the hair samples

exhibit the endothermal processes shown in Figures 5.9-5.10. Different heating rates

cause the endotherm peak to shift (Cao, 1997b). In agreement with Haly (1967), we

found that a substantial shift to higher temperature is observed in term of the

denaturation peak size in Figures 5.9 and 5.10 for both virgin hair and commercial

persulphate bleached (2h) hair with increasing heating rate.


DSC curves for virgin hair at various

heating rates. The raw data is given in the Chapter 8, Table 8.4.


DSC curves for 2h commercially

bleached hair at various heating rates. The raw data is given in the Chapter 8, Table 8.5.

-0.2

1.8

3.8

5.8

7.8

9.8

120 130 140 150 160

Cp N

R [

J/g

∙ oC

]

TD (oC)

0.5

1.5

2

2.5

3

4

5

7

β [oC/min]

8.5

9.5

10.5

11.5

12.5

13.5

110 120 130 140

Cp N

R [J/(g∙o

C)]

TD(oC)

0.5

1

1.5

2

3

4

5

7

β [oC/min]


124

Against the theoretical background given above, by measuring peak temperatures as a

function of the heating rate, the relationship between heating rate, β (as lnβ) and 1/TD

of denaturated hair fiber for various heating rates are investigated in a so-called

Arrhenius-plot. The advantages of the Arrhenius- plot method are simplicity of

measurement, applicability to many types of reactions and relative insensitivity to

secondary (side) reactions (Duswalt, 1974).

There are strong linear relationships between heating rate and 1/TD as well as ∆HD of

denaturated hair fibre in Figures 5.11 and 5.12. Activation energies are determined

from the slope of the linear regression between the non-reversing denaturation peak

temperature (as 1/TD) and heating rate, β (as lnβ), corresponding to 260 kJ/mol for the

virgin hair and 295 kJ/mol for the commercial persulphate bleached hair (2h). Both

values are in a good agreement with the range of activation energies which are

generally associates with protein denaturation (approx. 100-800 kJ/mol) (Bischof,

2005). In addition, Istrate et al. (2009) have verified that the kinetics are an

autocatalytic mechanism and the value of the activation energy is close to disulphide

bond scission rather than to protein denaturation.

Figure 5.11: The relationship between heating rate, β (as lnβ) and 1/TD for virgin and

commercial persulphate bleached hair (2h). Linear regression lines were determined by

the data from non-reversing heat capacity peaks only, yielding the values for the

activation energy EA, as given by Equation.5.19.

-2.3

-1.8

-1.3

-0.8

-0.3

0.2

0.7

1.2

1.7

2.2

0.00235 0.0024 0.00245 0.0025 0.00255 0.0026

lnβ

1/TD (1/K)

Virgin Hair

2h Commercially Bleached Hair

EA = 295 kJ/mol EA = 260 kJ/mol


125

The high value of 295 kJ/mol of the commercially bleached hair in the wet state (see

Figure. 5.11) reflects the high activation energy for chemical processes during keratin

denaturation, but the difference of activation energy between virgin hair and

commercial bleached hair is around 35KJ/mol, which is considered to be very small.

This conclusion indicates that there is no obvious difference for matrix properties

between virgin hair and commercially treated hair. This shows that the major structural

change produced by the treatment only occurs in the α-helical component in IF, which

is embedded into the matrix. The relationship of denaturation enthalpy ∆HD and

heating rate (Figure 5.12) provides a better explanation for this finding.

Linear regressions of ∆HD vs. lnβ (Figure 5.12) on both virgin hair and commercial

persulphate bleached hair (2h) show that there is a linear increase in ∆HD for lower

heating rates, while it is constant for higher heating rates. Again, this observation

suggests that higher heating rates favour the formation of a random coil structure,

while lower heating rates favour the formation of more organized, randomly oriented

small β-domains (Haly, 1970). If the heating rate is sufficiently small, the phase

change at the molecular level is from the α-helical structure to the β structure. On the

other hand, if the rate of heating is sufficiently high and cooling is rapid, little or no β

material is formed; in fact, there may be ‘recrystalization’ of material (Haly, 1970). So

it is taken as an indication that a lower heating rate favours a crystal transformation

change (α-β transformation), while a higher rate favours a crystalline-amorphous

transformation.

Figure 5.12: Relationship of denaturation enthalpy ∆HD and lnβ for virgin and

bleached hair

0

2

4

6

8

10

12

14

16

18

-3 -2 -1 0 1 2 3

∆H

D (

J/g)

lnβ

Virgin Hair

2h Commercially Bleached Hair


126

5.4.4.2 Nonisothermal Kinetics for the Denaturation of Mohair Fibres

Mohair is a very resilient hair fibre obtained from the Angora goat. In many respects,

mohair resembles wool in structure, including the characteristic scale structure of the

fibre. In term of its similar morphology and high keratin content, mohair is the closest

hair fibre to the human hair with regard to its general properties and structure with

respect to its thermal behaviour, which is reported by Tonin et al. (2004). In this

section, mohair is thus employed as a reference sample to provide specific information

about the nature and the course of the denaturation transition as well as the

denaturation mechanisms through kinetic thermal analysis.

Figures 5.13 and 5.14 show a typical DSC curves for mohair in water, after heating the

keratin at heating rates of 1oC/min and 3

oC/min, respectively. Though double peaks

were frequently observed for wool (Haly, 1967, Spei, 1989), the mohair sample shows

a single-peak structure for the denaturation. But there is a shoulder near the

denaturation temperature which appears at the lower heating rate towards higher

temperature in the Figure 5.14. This is a common feature for α-keratins (Wortmann,

1993) and is related to specific cell inhomogeneity of helix denaturation in wool

(Wortmann, 1998) and in human hair (Bryson, 2009), but the phenomena in mohair is

of special interest, because mohair is generally assumed to be homogenous with regard

to its cell structure.

It is not surprising that the virgin mohair sample demonstrates a similar DSC curve

shape as human hair, but the denaturation temperatures of the α-helical component in

mohair sample, shown at 133oC and 138

oC in Figures 5.13 and 5.14, respectively, are

slightly lower when compare to virgin human hair (138oC, 144

oC) at 1

oC/min and

3oC/min, respectively. This finding indicates that mohair has an overall similar α-helix

content as well as cross-link density in the matrix as the human hair.


127

Figure 5.13: Typical non-reversing Cp

NR DSC curve for untreated mohair hair in water,

when the heating rate is 1oC/min. The arrow shows the inhomogeneous cells

component on the DSC curve.



when the heating rate is 3oC/min


128

Table 5.8 provides the corresponding kinetics results for the mohair sample for the

ASTM E-698 kinetics approach by applying multiple heating rates.

Table 5.8: Results of mohair at various heating rates

Heating Rate

(oC/min)

0.1 0.2 0.3 0.5 1 2 3 4 5 7

Modulation

(oC/min)

0.05 0.1 0.1 0.1 0.5 1 1 1 1 3

∆T(K)

120.1

122.2

122.5

126.8

126.9

126.9

128.3

128.9

132.0

131.7

132.1

133.9

133.8

134.8

136.4

138.1

137.8

139.0

139.1

138.6

141.3

140.7

140.5

141.2

141.4

141.3

143.8

144.5

144.8

∆H(J/mol)

7.1

7.6

7.4

9.1

10.4

9.7

9.9

11.0

11.0

12.2

11.2

12.9

11.9

12.1

12.6

13.7

13.2

11.6

12.0

12.1

12.2

13.3

13.0

11.8

11.1

12.6

12.9

12.4

13.6

Figure 5.15 demonstrates the activation energy of 296 KJ/mol for the mohair sample,

which is determined from the slope of the regression lines of lnβ vs. 1/TD by the

Arrhenius-equation.

Figure 5.15: The relationship between heating rate, β (as lnβ) and 1/TD for mohair.

Linear regression lines were determined for data from non-reversing heat capacity

peaks only, yielding the values for the activation energy EA, as given by Equation. 5.19.


129

The heating rate effect on the mohair is reflected in the change of denaturation

enthalpy. As the heating rate increases, there is a linear increase in ∆HD for lower

heating rates, while it is effectively constant for higher rates in Figure 5.16, further

proved that the low heating rate has an extensive structural change in IF in the kinetic

analysis. The reason of this phenomenon has been explained in section 5.4.4.1.

Figure 5.16: Relationship of denaturation enthalpy ∆HD and lnβ for mohair

Furthermore, the linear regression of the mohair and virgin human hair are analysed

and compared in Figure 5.17. It is apparent that the two samples follow similar trends.

At the lower heating rates there is a linear increase in ∆HD, whereas at the higher

heating rate, the ∆HD remains constant. The slope of linear regression for the lower

heating rates increases from the mohair to virgin human hair, indicating that the

amount of β- structure, which convert from α-helical material is abundant in the virgin

mohair than in the virgin human hair. The values of denaturation enthalpies of mohair

and human hair overlap at the higher heating rate. From this it can be concluded that

both virgin keratin fibres have similar amounts of “native” α-helical material.

(Wortmann, 2002, 2006).

5

7

9

11

13

15

-2 -1 0 1 2

∆H

D(J

/g)

lnβ


130

Figure 5.17: Relationship of denaturation enthalpy ∆HD and lnβ for mohair and virgin

human hair

5.4.4.3 Morphological Investigation of Denaturated Human Samples by SEM

after DSC

Morphological Investigation of Denaturated Virgin Hair by SEM

SEM was employed to investigate the change of fibre appearance after the

denaturation process in DSC (Figure 5.18). The hair samples were collected from the

sample pans after DSC and dried at room temperature for the SEM investigation. All

of the experimental settings are the same as described in the Chapter 3.

Five different heating rates were utilized to illustrate the morphological changes of the

virgin hair by heating. Virgin hair has a smooth and intact cuticle edge (Fig.5.18 A),

after heating at a high heating rate (7oC/min) protein granules are observed on the hair

cuticle surface, which are assumed to be hydrolyzed proteins crystallized in the dry

state (Fig.5.18 B). The cuticle has the tendency to shrink at medium heating rates

(5oC/min) after denaturation of the α-helix in the IF structure (Fig.5.18 C). This is

explained by the fact that helix denaturation leads to supercontraction of the cuticle

appearance.

The phenomenon of cuticle distortion appears at a heating rate of 2oC/min, when

additional granules are found on the surface (Fig.5.18 D). As the process of

denaturation proceeds, the protein molecule unfolds and the internally directed

0

2

4

6

8

10

12

14

16

18

-2.5 -1.5 -0.5 0.5 1.5 2.5

∆H

D(J

/g)

lnβ

Mohair

Virgin human hair


131

hydrophobic regions in the molecule become exposed, so protein molecules aggregate

by non-polar, hydrophobic groups into a large insoluble mass (Istrate, 2009). At very

low heating rates (0.15oC/min), the cortex has been total dissolved and only the cuticle

remains (Fig.5.18 E), which confirms that hair fibre damage is stronger at lower

heating rates.

Figure 5.18: Cuticle fragment of virgin human hair fiber snippets by SEM after being

heated in the DSC at various heating rates up to 7oc/min.


132

Morphological Investigation of Denaturated Bleached Hair (2h) by SEM

The appearances of 2h commercially bleached hair fibres were also examined after the

DSC-experiments at different heating rates by SEM (Figure 5.19). 2h commercially

bleached hair has a smoother edge than virgin hair (Fig.19 A), at high heating rates

(7oC/min) only few protein granules are observed on the hair cuticle as hydrolyzed

proteins crystallize (Fig.19 B). The phenomenon of cuticle disintegration appears only

at medium rates (2oC/min) (Fig.19 D). At very low rates (0.15

oC/min) the cortex is

completely dissolved. Only cuticle remains (Fig.19 E).

In general, 2 h commercially bleached hair shows an overall shrunk cuticle appearance

and less protein granules than the virgin hair at the various heating rates. This

shrinkage phenomenon is primarily ascribed to that the denatured hair has a previously

partly destroyed α-helix in the cortical cell. These crystalline components appear to

exist the whole length of the cortical cell, thus the denaturation of the α-helices causes

a length change of the fibre.

In addition, the previously damaged commercial bleached hair contains the fewer

peptide and cystine bonds, thereby fewer and shorter protein granules appear on the

cuticle surface.


133

Figure 5.19: Cuticle fragment of commercial persulphate bleached human hair fiber

snippets (2h) after SEM after being heated in the DSC at various heating rates up to

7oc/min.


134

5.4.5 DSC Deconvolution

5.4.5.1 Background

As mentioned in section 5.1.3.3, the overlap of the denaturation peaks for the various

types of cortical cells in the human hair and mohair, results in the formation of a

shoulder in the denaturation peak on the high temperature side. The ortho- and para-

cortical cells have a different morphological structures, which, e.g. in Merino wool, are

arranged bilaterally in the crossed section. The orthocortex contains a lower

concentration of disulfide linkages than the para-cortex, thus showing a lower

denaturation temperature. In an aqueous medium, the denaturation temperature at

138°C is attributed to the ortho-cortical cells and that at 144°C is attributed to the para-

cortical cells. The phenomena of a bimodal peak is also observed at 230 and 240°C in

a dry environment, which it is again ascribes to the ortho- and para-cortical cells,

respectively (Wortmann, 1998). In this section, peak deconvolution is now employed

in order to resolve the shoulder denaturation peak of the keratin of human hair and

mohair, for the purpose of accurate thermal analysis.

5.4.5.2 Analytical Approach

Although the deconvolution process enables the overlapping peak to be resolved, the

results still have the possibility for bias. Wortmann and Deutz (1993, 1998) stated that

the assignment of peak temperatures became more accurate as peak analysis software

has improved. In order to explore the possibility that human hair and mohair may have

two peaks similar to those observed in wool fibres, a deconvolution was conducted

using an Excel spreadsheet and the “solver” function.

Our working hypothesis of the deconvolution of the peak overlapping was on the basis

of an observation of Freire (1978) that the heat flow–temperature curves can be

decomposed into the sum of Gaussian functions. The Gaussian function is a function

of the form:


135

(5.22)

The graph of a Gaussian is a characteristic symmetric “bell curve” shape that quickly

falls off towards plus/minus infinity. The parameter α is the height of the curve’s peak,

b is the position of the centre of the peak, and c controls the width of the “bell”.

The initial DSC scans of human hair and mohair were produced by the TA software.

They then were transferred into an established Excel program sheet to achieve a DSC

curve. The first step of the deconvolution is to correct the shifted baseline of DSC

curve prior to the decomposition into the Gaussian curves. In principal, the slope of the

chosen points on the original DSC curve is identical to the slope which the straight line

is determined by the TA software. Figure 5.20 demonstrates the process of baseline

correction. A new baseline with its own slope is applied between two chosen points on

the original curve and the data on the previous baseline accordingly are converted with

the new baseline, therefore creating a DSC curve with a new perfectly horizontal

baseline.

Figure 5.20: The DSC baseline of virgin human hair between the two chosen points, at

a heating rate of 3oC/min.

10.5

11

11.5

12

12.5

13

120 130 140 150 160

Cp

NR (

J/g)

Temperature (oC)


136

Figure 5.21 represents the corrected DSC peak, which is achieved by subtracting the

CpNR

with its new linear baseline slope. Then this peak is deconvoluted using a set of

three peaks according to Equation.5.22 in the Excel program (Figure 5.22). The

parameter for three peaks were determined by this procedure and attributed to para-,

ortho-, and background- cortical cells (see Figure 5.22).

Figure 5.21: The corrected DSC peak of virgin human hair, at a heating rate of

3oC/min

Figure 5.22: Deconvolution of the MDSC-curve for virgin human hair into three

Gaussian distributions. The peaks are assigned to the three types of cell types as para-,

ortho- and background cells. The dots (●) indicate the absolute differences between fit

and experimental data. The heating rate is 3oC/min.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

125 135 145 155

Cp

NR

(J

/g)

Temperature (oC)

Corrected Peak

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

125 135 145 155

Cp

NR

(J

/g)

Temperature (oC)

Background,

140.7oC, 31.1%

Ortho, 143.5oC,

45.6%

Para, 147.3oC,

23.2%


137

In Figure 5.22, a deconvolution of the DSC-peak for virgin human hair is shown for

three Gaussian distributions, attributed to para-, ortho- and background- cells. Their

peak temperatures and fractional enthalpies are given on the graph. According to the

ortho/para hypothesis, the first peak at 143.5°C corresponds to an orthocortex and the

second peak at 147.3°C is designated as paracortex because of its high cystine content.

Its fractional enthalpy (%) corresponds to ortho- (45%) and para- cortical (23%) cells,

respectively. It has been suggested that the DSC bimodal endotherm peaks are 4-7 oC

apart in the presence of ortho- and para-cortical cells for wool and other keratins in

water (Wortmann, 1998), the difference between the peak temperature (3.8oC) is in

satisfactory agreement with expectations for ortho- and para-cells.

The broad background peak (31%) summarizes the enthalpy fraction, which represents

the sizeable fraction of cells, but it is not clearly associated with the two distinct cell

types. The peak is assumed to comprise the small fractions of strongly staining (high

sulfur) as well as weakly staining cells (low sulfur) in Bryson et al.’s (2009)

microscopic observations. Since the background peak temperature (140 oC) is lower

than ortho and para, it is unlikely that it refers to meso-cortical cells (Whiteley, 1977),

but in our investigation the background peak is attributed to meso-cell peak to avoid the

confusion.

The various set of raw data obtained from Section 5.4.3, which investigates the

damage to the cortex for various bleaches and the raw data obtained from Section 5.4.4

on the kinetic analysis using various heating rates for virgin human hair and mohair,

were all transferred into the Excel program sheet, by using a deconvolution program

with a three peak assumption. The results are present and summarized in Tables 5.9-

5.14.

5.4.5.3 Results and Discussion

Deconvolution Investigation of Human Hair on the Various Bleaches Effects

In the case of keratin fibres, allocating the lower peak to the ortho- and the higher one

to the para-cortex, the peak temperatures were determined and are summarized,

together with the denaturation enthalpy as percentage fraction of the peak area. Tables


138

5.10—5.12 summarize the deconvolution results for the peak temperatures TD and

fractional enthalpies (%) for the cell types for 6% H202 bleached hair, 9% H202

commercial bleached hair and commercial persulphate bleached hair, respectively, at a

heating rate of 3 oC/min.

For the comparison purpose among the three bleaches, the 6% H202 bleached sample

(2h) is taken as an example. With regard to the denaturation enthalpy of ortho- and

para-cortical cells, it is obvious that the higher value of the para- cortex (56%)

compared to the ortho-cortex (32%) in the 6% bleached hair is well meets the

expectation, which is based on the investigations by Dobb (1970), that the ortho-cortex

should contain about double the amount of α-helical material than the para-cortex. But

the 9% H202 commercial bleached hair (2h) and commercial persulphate bleached hair

(2h) behaves irregularly. In both cases the enthalpy fraction attributed to the para-

cortical cells is reduced, while the ortho-cortex appears largely unchanged. This needs

further investigation.

Table 5.9: Peak temperatures and fractional enthalpies for para-, ortho- and meso-

cortical cells for 6% H202 bleached hairs at various bleaching times, in the form of the

arithmetic means, standard error is given in the Figure 5.23.

Bleaching Time

Peak Temperature

0 0.5h 1h 1.5h 2h 2.5h 3h

Para 147.1 142.8 141.2 142.5 140.9 140.8 140.4

Ortho 143.9 139.3 138.0 138.6 138.6 138.1 138.2

Meso 140.7 133.4 131.3 132.4 133.2 133.1 133.1 Bleaching Time

Fractional Enthalpy (%) 0 0.5h 1h 1.5h 2h 2.5h 3h

Para 31.0 62.3 57.6 66.3 56.7 61.8 53.9

Ortho 44.4 29.8 38.8 29.8 32.2 26.7 37.7

Meso 24.6 8.0 3.6 4.0 11.2 11.5 8.5


139


cortical cells for 9% H202 commercial bleached hairs at various bleaching times, in the

form of the arithmetic means, standard error is given in the Figure 5.24.

Bleaching Time

Peak Temperature 0 0.5h 1h 1.5h 2h 2.5h 3h

Para 147.1 141.8 140.7 139.1 137.1 133.1 136.1

Ortho 143.9 140.1 138.3 135.6 133.1 127.3 130.6



Para 31.0 33.8 26.9 27.0 14.0 49.6 26.7

Ortho 44.4 47.9 35.6 35.4 34.4 32.7 39.6

Meso 24.6 37.5 37.6 51.7 17.8 33.4 8.5


cortical cells for commercial persulphate bleached hairs at various bleaching times, in

the form of the arithmetic means, standard error is given in the Figure 5.25.

Bleaching Time

Peak Temperature 0 0.5h 1h 1.5h 2h 2.5h 3h

Para 147.1 138.1 130.6 128.2 126.4 127.3 126.1

Ortho 143.9 135.6 127.6 124.9 122.7 124.2 123.0



Para 31.0 30.5 32.0 39.3 38.3 33.1 28.8

Ortho 44.4 54.6 39.1 34.8 34.7 35.8 38.1

Meso 24.6 14.9 28.8 25.9 27.0 31.1 33.0

It was speculated previously that the 9% H202 commercial bleach shows an irregular

damage profile compared to 6% H202 bleach and commercial persulphate bleach after

2h bleaching time. Therefore, the denaturation temperatures of the resolved cortical

compositions of 6% H202 and commercial persulphate bleached hairs are compared.

The results are presented in Figure 5.23. It is obvious that commercial persulphate

bleached hair has an overall more progressive damage influence on ortho- and para-

cortical cell compositions than 6% H202 bleached hair. The denaturation temperatures

decrease faster for the first 2h bleaching time until reaching an equilibrium level. On

the other hand, 6% H202 bleach has a relatively gentle oxidation impact, although it


140

shows a big drop for denaturation temperature for the initial 1h treatment time. Its

denaturation temperature still has a higher value than for the commercial persulphate

bleached hair. This observation indicates that the para- and ortho- cortex of 6% H202

bleached hair are subject to little structural change compared to commercial

persulphate bleached hair, thus leading to higher denaturation temperatures.

Figure 5.23: Denaturation temperatures of para-, ortho- and meso- cortical cell types

on the basis of three Gaussian distributions for 6% H202 and commercial persulphate

bleached hairs up to 3hr bleaching time.

The three denaturation temperatures for 9% H202 commercial bleach are analyzed in

Figure 5.24. TD decreases linearly over the whole reaction time for all of the three cell

groups.

115

120

125

130

135

140

145

150

0 1 2 3

De

nat

ura

tio

n T

emp

erat

ure

(oC

)

Bleaching Time (h)

6% Bleach Para

6% Bleach Ortho

6% Bleach Meso

Commercial Para

Commercial Ortho

Commercial Meso


141


on the basis of three Gaussian distributions for 9% H202 commercial bleached hairs up

to 3hr bleaching time.

Consequently, it can be seen from the Figures 5.23 and 5.24 that the commercial

persulphate bleach has a strong damage effect on both para- and ortho- cortex. And 6%

H202 bleach has an overall mild structural impact on both two main cell types. This

phenomenon confirms our previous finding in Section 5.4.3 that commercial

persulphate bleach and 6% H202 bleach approached their equilibrium within the 3hr

treatment time, while 9% H202 commercial bleach shows a linear decrease on the

structure change with the increasing time.

The denaturation enthalpies are also compared. The comparison of the three cortical

cell types in the form of their fractional percentages for 6% H202 bleached hair is

shown in Figure 5.25. The phenomenon reflects that the 6% H202 bleach has a bigger

damage effect on the ortho-cortex rather than the para-cortex. However, the values of

the resolved cortical cells are uniformly distributed along the treatment time, further

confirms that 6% H202 bleach rather is the mild oxidising bleaching chemical among

the three bleaches.

120

125

130

135

140

145

150

0 0.5 1 1.5 2 2.5 3

De

nat

ura

tio

n T

emp

erat

ure

(oC

)

Bleaching Time (h)

9% H202 Commercial Para

9% H202 Commercial Ortho

9% H202 Commercial Meso


142

Figure 5.25: Fractional enthalpies of para-, ortho- and meso- cortical cell types on the

basis of three Gaussian distributions for 6% H202 bleached hairs up to 3hr bleaching

time.

The data of 9% H202 commercial bleached hair show an irregular pattern in Figure 5.26,

both para- and ortho-cortical cells have a downward trend, and they seem to approach

similar values of fractional enthalpy with the longer oxidation time, reflecting that 9%

H202 commercial bleach has a homogenous damage effect on the para-cortex and ortho-

cortex.


basis of three Gaussian distributions for 9% H202 commercial bleached hairs up to 3hr

bleaching time.

0

10

20

30

40

50

60

70

80

0 1 2 3

Frac

tio

nal

En

thal

py

(%)

Bleaching Time(h)

6% H202 Bleach Para

6% H202 Bleach Ortho

6% H202 Bleach Meso

0

10

20

30

40

50

60

70

80

0 1 2 3

Frac

tio

nal

En

thal

py

(%)

Bleaching Time (h)

9% H202 Commercial Para

9% H202 Commercial Ortho

9% H202 Commercial Meso


143

In Figure 5.27, the value for the ortho-cortex in the commercial persulphate bleached

hair keeps decreasing while both para- and meso- cell compositions increase until the

values for all cortex components overlap at around 1.5hr bleaching time, which is in the

range of around 30% fractional enthalpy. This observation is also found for 9% H202

commercial bleached hair, consequently leading to the conclusion that the stronger

bleaches result in a homogenous structural damage on both para-cortex and ortho-

cortex.


basis of three Gaussian distributions for commercial persulphate bleached hairs up to

3hr bleaching time.

Deconvolution Investigation of Human Hair for Different Heating Rates

The overall damage effects on intermediate filament and matrix through the

commercial persulphate bleached hair (2h) at various heating rates were discussed in

Section 5.4.4.4. In this investigation the deconvolution is utilized to further evaluate the

oxidation effect on the different cortical cells (mainly the orthocortex and paracortex).

Since the paracortex has a higher sulphur content, the higher peak temperature is

attributed to the para- rather than the orthocortex. In the Table 5.12, the value for virgin

human hair of 147oC corresponds to the paracortex, and 143

oC is attributed to the

orthocortex component, for the heating rate of 3oC/min.

10

15

20

25

30

35

40

45

50

55

60

0 1 2 3

Frac

tio

nal

En

thal

ply

(%

)

Bleaching Time(h)

Commercial Persulphate Para

Commercial Persulphate Ortho

Commercial Persulphate Meso


144

Table 5.12: Peak temperatures and fractional enthalpies of three cell types in virgin

hair at various heating rates, in the form of the arithmetic means, standard error is

given in the Figure 5.28.

Heating rate(oC/min)

Peak Temperature 1.5 2 2.5 3 4 5 7

Para 140.4 142.4 142.8 147.1 148.9 150.4 150.8

Ortho 137.6 139.5 140.3 143.9 145.4 147.2 147.1

Meso 134.1 136.6 135.6 140.7 141.6 143.8 143.2 Heating rate(

oC/min)

Fractional Enthalpy(%) 1.5 2 2.5 3 4 5 7

Para 33.0 34.0 43.0 31.0 38.0 45.3 37.9

Ortho 37.0 32.9 41.4 44.4 39.1 29.7 37.1

Meso 30.0 33.1 15.6 24.6 22.9 25.0 25.0

The enthalpy is also investigated in the form of the fractional enthalpies in Tables 5.12

and 5.13 for virgin and persulphate bleached hairs. It can be seen that virgin hair and

commercial persulphate bleached hair (2h) have shown three distinctive cell

compositions after the heating effects, and commercial persulphate bleached hair (2h)

have an overall lower denaturation temperature for three cell compositions than the

virgin hair.

Table 5.13: Peak temperatures and fractional enthalpies of three cell types in

commercial persulphate bleached hair (2h) at various heating rates, in the form of the

arithmetic means, standard error is given in the Figure 5.29.

Heating rate(oC/min)

Peak Temperature 1 1.5 2 3 4 5 7

Para 121.7 123.4 124.8 127.6 128.6 130.0 132.7

Ortho 117.8 119.9 120.6 123.8 124.5 125.7 127.9

Meso 112.6 115.8 115.7 119.4 120.0 121.0 122.4 Heating rate(

oC/min)

Fractional Enthalpy(%) 1 1.5 2 3 4 5 7

Para 43.5 47.8 54.5 47.9 49.1 49.3 46.1

Ortho 37.0 39.2 39.5 38.3 35.6 35.7 35.9

Meso 19.6 13.1 6.0 13.8 15.2 15.0 18.1


145

A continuous increase in the denaturation temperature with the increasing heating rates

is observed for all three cell types in virgin hair as well as commercial persulphate

bleached hair (2h), as shown in Figure 5.28. The result reflects well the finding for the

whole human hair in the Section 5.4.4.1. Although commercial persulphate bleached

hair (2h) has an overall lower denaturation temperature for each cortical component for

each heating rate, the distance between the values for para- and ortho- cortex for virgin

hair is similar to that of commercial persulphate bleached hair (2h). This shows that

there is a parallel change in the denaturation temperature for virgin hair and

commercial persulphate bleached hair (2h). It further leads to the conclusion that

heating rate affects the cystine linkages of the para- and ortho-cortex homogenously,

regardless of the previous damage.


on the basis of three Gaussian distributions for virgin and commercial persulphate

bleached hairs (2h) at various heating rates.

However, there is no strong relationship between the fractional enthalpies for virgin

hair and heating rate as shown in Figure 5.29, this is ascribes to the disulfide cross-link

protection of the crystalline component. Commercial persulphate bleached hairs show a

uniform performance for the three types of cells, indicating that a homogenous damage

effect on the cortical components occurs for bleached hair.

100

110

120

130

140

150

160

1 2 3 4 5 6 7

De

nat

ura

tio

n T

em

per

atu

re (

oC

)

Heating Rate ( oC/min)

Virgin Para Virgin Ortho

Virgin Meso Commercial Para

Commercial Ortho Commercial Meso


146


basis of three Gaussian distributions for virgin and commercial persulphate bleached

hairs (2h) at various heating rates.

Deconvolution investigation of Mohair on the kinetic analysis

Mohair, which has the highest α-helix content among the animal fibres contains

predominantly ortho-cortex, or meso- and orthocortex (Jones, 2006). But kid mohair

also contains some para-cortex (Hunter, 2001). In the section 5.4.4.2, it had been

found that there is a shoulder in the DSC-curve on the high temperature side at the

heating rate of 1 oC/min in water. In this section 5.4.4.3, deconvolution is employed to

investigate whether cell composition of mohair can be resolved into the three Gaussian

distributions.

Separate denaturation temperatures have been found for the three different cell types:

para-, ortho and meso-cortex (see Table 5.14). When the heating rate is 3oC/min,

according to the ortho/para hypothesis, the first peak at 136.3°C corresponds to

orthocortex and the highest peak at 140.5°C is designated to paracortex, because of its

high cystine content. The middle peak (138.2°C), which is intermediate between para-

and ortho-cortex, is assigned to meso-cortex. Their fractional enthalpies (%) have been

found to correspond to ortho- (34.3%), para- (50.4%) and meso-cortical (15.3%) cells,

respectively.

0

10

20

30

40

50

60

1 2 3 4 5 6 7

Frac

tio

nal

En

thal

py

(%)

Heating Rate (oC/min)

Virgin Para

Virgin Ortho

Virgin Meso

Commercial Para

Commercial Ortho

Commercial Meso


147

Table 5.14: Peak temperatures and fractional enthalpies of para-, ortho- and meso-

cortical cells in virgin mohair for various heating rates, in the form of the arithmetic

means, standard error is given in the Figure 5.30.

Heating rate (oC/min)

Peak Temperature 1 2 3 4 5 7

Para 136.2 139.1 140.5 142.0 142.1 144.9

Ortho 131.0 135.4 136.3 137.6 139.2 138.6

Meso 133.4 136.7 138.2 139.5 139.6 140.7 Heating rate(

oC/min)

Fractional Enthalpy(%) 1 2 3 4 5 7

Para 41.1 48.8 50.4 60.1 73.6 75.0

Ortho 28.5 38.9 34.3 25.7 16.0 14.8

Meso 30.4 12.3 15.3 14.1 10.5 10.2

The denaturation temperatures of three cell types in mohair increase continuously with

the heating rate (Figure 5.30), but para-cortex shows a more dominant increase. This is

in contrast to the fact that mohair has predominately ortho-cortex content than para-

cortex. The reason for this abnormal result may be ascribed to (i) the randomly

collected sample source, or (ii) the deconvolution with a “Excel sheet” program is not

an appropriate analysis approach for this keratin fibre, which has only one or two types

of cell cortex.

Figure 5.30: Denaturation temperature for para-, ortho- and meso- cortical cell

fraction on the basis of three Gaussian distributions for virgin mohair at various

heating rates.

130

132

134

136

138

140

142

144

146

1 2 3 4 5 6 7

De

nat

ura

tio

n T

emp

erat

ure

(oC

)


Para-Cortex Ortho-Cortex Meso-Cortex


148

The fractional denaturation enthalpies of the three cell types are given in Figure 5.31.

For both ortho- and meso- cortex the values decrease apparently with the higher

heating rate while the para-cortex shows a steady upward tendency over the heating

range. However, this observation is unexpected.

Figure 5.31: Fractional enthalpies for para-, ortho- and meso-cortical cell fraction on

the basis of three Gaussian distributions for virgin mohair at various heating rates.

5.5 Morphological Changes on Three Ethnics Hair Types

after DSC investigations

5.5.1 Theory and Background

Ethnicity is an important factor in determining the formulation of cosmetic products in

the industry. Human hair is categorised into 3 major distinct groups according to ethnic

origin: Asian, Caucasian and African. Different ethnic hair types are significantly

different in their structure and properties. The distinction between these hair types is

particularly related to diameter, geometry, crimp and colour.

Asian hair is found to be the strongest and thickest followed by Caucasian hair.

African hair is generally more liable to damage and breakage, and has a lower moisture

content than Asian or Caucasian hair. Asian hair is generally very straight, Caucasian

0

10

20

30

40

50

60

70

80

90

1 2 3 4 5 6 7

Frac

tio

nal

En

thal

p(%

)


Para-Cortex Ortho-Cortex Meso-cortex


149

hair varies from straight to wavy to moderately curly, in contrast with the African hair

which exhibits a characteristic tightly curled structure (Wei, 2005). These variations in

the shape of hair from different ethnic origins depend on several factors, including the

shape of the hair follicle and its opening; these vary from one person to another and

also between races (Gray, 2001, Thibaut, 2005b). But all hair, irrespective of ethnic

origin, have the same protein structure and composition (Menkart, 1984, Dekio, 1988b,

Dekio, 1990a).

In this study, the morphology changes on three ethnic hair types are compared using

SEM, which have been heated at different isothermal heating times by DSC

measurement.

5.5.2 Experimental

5.5.2.1 Hair Samples

Caucasian, Asian and African hair samples were used in this study. The hair samples

were collected from volunteers. Although the exact location from the root is unknown,

it is estimated that hair samples used for testing were taken at 10-20 cm from the scalp.

In most cases, the experiments were conducted on the middle parts of the hair samples

(about 10 mm).

5.5.2.2 DSC Experimental Work

All the experimental settings are the same as described in Section 5.4.3. The standard

modulated DSC mode (MDSC) was applied, and hairs from three different ethnic

origins were heated with the temperature increasing linearly at a rate of 5oC/min in the

range of 80-180oC. Thus denaturation temperature and denaturation enthalpy can be

determined, as illustrated in Figure 5.32. Because there is no distinct difference

between three types of hair, their denaturation temperature is observed at a similar

location (~147oC).


150


DSC curve for three types of human hair in

water at a heating rate of 5oC/min

In the next step, the virgin hair samples were heated with the temperature increasing

linearly at a rate of 5oC/min up to the peak temperature of 147

oC, and left for 5, 10 and

20mins of isothermal time, individually, before cooling down. This procedure was

chosen to examine the hydrolytic influence on hair morphology at their denaturation

temperature (147oC).

5.5.2.3 SEM Experimental work

The residues of denaturated hair samples after DSC were collected and dried at room

temperature for the further SEM investigation. All the experimental settings are as

same as described in Chapter 4. The sample description for SEM images is given in

Table 5.15.

10

10.5

11

11.5

12

12.5

13

13.5

14

14.5

120 130 140 150 160 170 180

Cp

NR

J/

(g∙o

C)

Temprature (oC)

Asian Hair

Caucasian Hair

African Hair

147oC


151

Table 5.15: Description of SEM images for various hair samples

Picture No. Ethnicity Isothermal Time(mins)

A Asian 5

B 10

C 20

D African 5

E 10

F 20

G Caucasian 5

H 10

I 20

5.5.3 Results and Discussion

SEM images in Figure 5.33 are employed to document the changes of the

morphological structures of the three types of hair after being heated at various

isothermal times. The cuticle of Asian hair has a stronger tendency towards distortion.

Numerous hydrolysed proteins granules appear with longer isothermal heating time

(Pictures 5.33 A-C). The images also have electron dense bodies visible in the cuticle

(red arrow), these are consistent in size and shape (approximately 1–5 mm) with

melanosomes. Since melanin granules are rarely present in the cuticle (Robbins, 2002),

it is possibly an artifact of the contamination process (Stout, 2007) or of the large

amount of melanin granules in Asian hair, which has a natural black colour.

African hair, which maintains as smooth and intact cuticle surface as the virgin

Caucasian hair, shows a very high thermal resistance (Pictures 5.33 D-F), this may be

ascribed to its hardness and the curly properties of the hair nature.

Caucasian hair samples demonstrate a similar phenomenon in Pictures 5.33 G-I, as the

virgin Caucasian hair in the kinetic analysis in Chapter 5.4.4.3. The hydrolysed protein

granules increase with increasing isothermal heating times, which implies that IFs have

been progressively destroyed by heating.


152

B: 10 mins A: 5 mins

C: 20mins D: 5mins

E: 10mins F: 20mins

MINS


153

Figure 5.33: SEM images of three types of hair after being heated to their denaturation

temperature and left at various isothermal times, before cooling. The image

identifications are described in Table 5.14.

5.6 Conclusion

DSC analyses confirm the effect of bleaching on hair damage. Since ∆HD and TD

reflect the amount and structural integrity of the α-helical material in the IFs, and

respectively the cross-link density of the matrix (IFAPs), in which the IFs are

embedded (Wortmann, 2002), there is a consistent decrease in both denaturation

temperatures TD and enthalpies ∆HD over the bleaching time. This shows that the longer

the bleaching time, the greater is the structural damage produced in hair. Thus a lower

keratin denaturation enthalpy and a lower denaturation temperature are observed.

Consequently the commercial persulphate bleach has a stronger oxidation effect on

both intermediate filaments and matrix than 9% H202 commercial bleach and 6% H202

G: 5mins H:10mins

I: 20mins


154

bleach. The deconvolution results of the DSC-curves for the three bleaches confirm this

observation. Stronger bleach results in similar damage for the crystalline component in

both para- and ortho-cortical cells with prolonged bleaching time.

However, a strong linear relationship was observed for 6% H2O2, commercial

persulphate bleach and part of 9% H202 commercial bleach by plotting ∆HDR

against

TDR, suggesting that the three bleaches have a homogenous bleaching effect on IFs and

IFAPs. However, 9% H202 commercial bleached hair after 2h bleaching time is

considered that the intermediate filament is subjected to progressive destruction after

excessive bleaching treatment, thus causing unfavourable results for the 9% H202

commercial bleached hairs.

SEM examined the appearance changes of virgin and bleached hair after keratin

denaturation at various heating rates (kinetic analysis). It is concluded that lower

heating rates result in more severe structural changes. The commercial persulphate

bleached hairs show an overall shrunk cuticle and less protein granules, due to the

previously denaturated α-helix in the cortical cells. In addition, the deconvolution

results further support that heating rates effect affects the cystine linkages in the para-

and ortho-cortex homogenously, regardless of the previous damage. Commercial

persulphate bleached hair shows a uniform damage effect for para-, ortho- and meso-

cortical components.

5.7 References

ALBERTS, B., BRAY, D., JOHNSON, A., LEWIS, N., RAFF, M., ROBERT, K.,

WALTER, P., 1998. Essential Cell Biology: An Introduction to the Molecular Biology

of the Cell, New York, Garland Publishing, Inc.

BISCHOF, J. C., HE, X., 2005. Thermal Stability of Proteins. Annals of the New York

Academy of Sciences, 1066, 12-33.


155

BISCHOF, J. C., WOLKERS, W. F., TSVETKOVA, N. M., OLIVER, A. E.,

CROWE, J. H., 2002. Lipid and protein changes due to freezing in Dunning AT-1 cells.

Cryobiology, 45, 22-32.

BORCHARDT, H. J., DANIELS, F., 1957. The Application of Differential Thermal

Analysis to the Study of Reaction Kinetics. Journal of the American Chemical Society,

79, 41-46.

BRYSON, W. G., HARLAND, D.P., CALDWELL, J.P., VERNON, J.A., WALLS,

R.J., WOODS, J.L., NAGASE, S., ITOU, T., KOIKE, K., 2009. Cortical cell types and

intermediate filament arrangements correlate with fiber curvature in Japanese human

hair. J. Struct. Biol, 166, 46-58.

BUNJES, H., UNRUH, T., 2007. Characterization of lipid nanoparticles by differential

scanning calorimetry, X-ray and neutron scattering Advanced Drug Delivery Reviews,

59, 379-402.

BURNHAM, A. K. 2000. Computational aspects of kinetic analysis. Part D: The

ICTAC kinetics project -multi-thermal-history model-fitting methods and their relation

to isoconversional methods. Thermochimica Acta, 355, 165-170.

BURNHAM, A. K., BRAUN, R. L., 1999. Global Kinetic Analysis of Complex

Materials. Energy & Fuels,, 13, 1-22.

CAO, J. 1997a. Origin of the Bimodal "Melting" Endotherm of a-Form Crystallites in

Wool Keratin. Journal OF Applied Polymer Science, 63, 411-415.

CAO, J. 1999. Melting study of the a-form crystallites in human hair keratin by DSC.

Thermochimica Acta, 335, 5-9.

CAO, J., JOKO, K., COOK, J. R., 1997b. DSC studies of the melting behavior of a-

form crystallites in wool keratin. Textile Research Journal, 67, 117-123.

CAO, J., LEROY, F., 2005. Depression of the melting temperature by moisture for α-

form crystallites in human hair keratin. Biopolymers, 77, 38-43.

CHIEW, Y. C., KUEHNER, D., BLANCH, H. W., PRAUSNITZ, J. M., 1995.

Molecular Thermodynamics for Salt-Induced Protein Precipitation. AIChE Journal, 41,

2150-2159.


156

COOPER, A. 1999. Thermodynamics of protein folding and stability. In: ALLEN, G.

(ed.) Protein: A Comprehensive Treatise. Stamford, CT: JAI Press.

CRAVALHO, E. G., TONER, M., GAYLOR, D. C., LEE, R. C., 1992. Response of

cells to supraphysiological temperatures: Experimental measurements and kinetic

models. In: LEE, R. C., CRAVALHO, E. G., BURKE, J. F., (ed.) Electrical

Trauma:The Pathophysiology, Manifestations and Clinical Management. Cambridge:

Cambridge University Press.

CREIGHTON, T. E. 1990. Protein folding. Biochem J, 270, 1-16.

CRIGHTON, J. S. The characterisation of destabilised and stablised wools.

Proceeding of the 8th International Wool Textile Research Conference, 1990

Christchurch, New Zealand.

CRIGHTON, J. S., HOLE, E. R.,. A study of wool in aqueous media by high pressure

differential analysis. 7th International Wool Textile Research Conferenace, 1985

Tokyo, Japan. 283-292.

DEKIO, S., JIDOI, J., 1988. Hair low-sulfur protein composition does not differ

electrophoretically among different races. J Dermatol. , 15, 393-6.

DEKIO, S., JIDOI, J., 1990. Amounts of fibrous proteins and matrix substances in

hairs of different races. J Dermatol., 17, 62-4.

DOBB, M. G. 1970. Electron-diffraction Studies of Keratin Cells. Journal of the

Textile Institute, 61, 170.

DUSWALT, A. A. 1974. The practice of obtaining kinetic data by differential

scanning calorimetry Thermochimica Acta, 8, 57-68.

E2041 Estimating Kinetic Parameters by Differential Scanning Calorimetry Using the

Borchardt and Daniels Method. West Conshohocken, PA: ASTM International.

E2070 Kinetic Parameters by Differential Scanning Calorimetry Using Isothermal

Methods. West Conshohocken, PA ASTM International,.


157

EYRING, H. 1974. Temperature. In: JOHNSON, H. F., EYRING, H., STOVER, B. J.,

(ed.) The Theory of Rate Processes in Biology and Medicine. New York: John Wiley

& Sons.

FERNANDEZ D’ARLAS, B., RUEDA, L., STEFANI, P. M., DE LA CABA, K.,

MONDRAGON, I., ECEIZA, A., 2007. Kinetic and thermodynamic studies of the

formation of a polyurethane based on 1,6-hexamethylene diisocyanate and

poly(carbonate-co-ester)diol. Thermochimica Acta, 459, 94-103.

FEUGHELMAN, M. 1959. A Two-Phase Structure for Keratin Fibers. Textile Reseach

Journal, 29, 223.

FEUGHELMAN, M. 1989. A note on the water impenetrable component of alpha-

keratin fiber. Textile Res. J., 59, 739-742.

FEUGHELMAN, M. 1997. Mechanical Proeprties and Structure of Alpha-Keratin

Fibres: Wool, human hair and related fibres, University of New South Wales Press.

FLYNN, J. H., WALL, L. A., 1966a. General Treatment of the Thermogravimetry of

Polymers Journal of Research of the National Bureau of Standards - A Physics and

Chemistry, 70A, 487-523.

FLYNN, J. H., WALL, L. A., 1966b. A Quick, Direct Method for the Determination of

Activation Energy from Thermogravimetric Data. Polymer Letters, 4, 323-328.

FRANBOURG, A., LEROY, F., LEVEQUE, J. L., DOUCET, J., 1996. Synchroton

light: A powerful tool for the analysis of humanh air damage. 10th Int. Hair-Science

Symp. Rostock: German Wool Res. Inst.

FREIRE, E., BILTONEN, R. L., 1978. Statistical mechanical deconvolution of

thermal transitions in macromolecules. I. Theory and application to homogeneous

systems. Biopolymers, 17, 463-479.

GILL, P. S., SAUERBRUNN, S. R., READING, M., 1993. Modulated Differential

Scanning Calorimetry. Journal of Thermal Analysis, 40, 931-939.

GRAY, J. 2001. Hair Care and Hair Care Products. Clinics in Dermatology 19, 227-

236.


158

GROENEWOUD, W. M. 2001. Characterisation of Polymers by Thermal Analysis,

Amsterdam, Elsevier Science.

HALY, A. R., SNAITH, J. W., 1967. Differential thermal analysis of wool: the phase-

transition endotherm under various conditions. Textile Research Journal, 37, 898-907.

HALY, A. R., SNAITH, J.W., 1970. The Heat of the Phase Transformation in Wool

Keratin Under Various Conditions. Textile Research Journal, 40, 142.

HORIO, M., KONDO, T., SEKIMOTO, K., FUNATSU, M., 1965. Pyrolysis of Wool

in Sealed Tubes. Proceedings of 3rd International Wool Textile Research Conference

Paris.

HUMPHRIES, W. T., MILLER, D. L., WILDNAUER, R. H., 1972. The

Thermomechanical Analysis of Natural and Chemically Modified Human Hair.

Journal of the Society of Cosmetic Chemists, 23, 359-370.

HUNTER, L., HUNTER, E.L 2001. Appendix 3 Composition of mohair fibres and of

amino acids, Cambridge, England, Woodhead Publishing Ltd.

ISTRATE, D., POPESCU, C., MÖLLER, M., 2009. Non-Isothermal Kinetics of Hard

a-Keratin Thermal Denaturation. Macromolecular Bioscience, 9, 805-812.

JOLY, M. 1965. A Physico-chemical Approach to the Denaturation of Proteins,

London, Academic Press.

JONES, L. N., RIVETT, D. E., TUCKER, D. J., 2006. Wool and Related Mammalian

Fibers, Taylor & Francis Group, CRC Press.

KANETAKA, S., TOMIZAWA, K., IYO, H., NAKAMURA, Y., 1993. The effects of

UV radiation on human hair concerning physical properties and fine structure of

protein. J Soc Cosmet Chem Jpn, 27, 424-431.

KONDA, A., TSUKADA, M., KURODA, S., 1973. The change of thermal property

of wool keratin with drawing. Journal of Polymer Science: Polymer Letters Edition, 11,

247-251.

KUZUHARA, A. 2006. Aanalysis of structural changes in bleached keratin fibers

(black and white human hair) using Raman spectroscopy. Biopolymers, 81, 506-514.


159

KUZUHARA, A. 2007. Analysis of structural changes in permanent waved human

hair using Raman spectroscopy. Biopolymers 85, 274-283.

LEROY, F., FRANBOURGA, A., LEVEQUE, J. L., 1992. Thermoanalytical

investigations of reduced hair. 8th lnt. Hair-Scienc Symp. Kiel: German Wool Res. Inst.

LI, C. R., TANG, T. B., 1999. A new method for analysing non-isothermal

thermoanalytical data from solid-state reactions. Thermochimica Acta, 325, 43-46.

LI, C. R., TANG, T. B., 1997. Dynamic Thermall Analysis of Solid-state Reactions

Journal of Thermal Analysis, 49, 1243-1248.

LUMRY, R., EYRING, H., 1954. Conformation Changes of Proteins. J. Phys. Chem,

58, 110-120.

LYUBAREV, A. E., KURGANOV, B. I., 2000. Analysis of DSC data relating to

proteins undergoing irreversible thermal denaturation Journal of Thermal Analysis and

Calorimetry, 62, 49-61.

MACIEJEWSKI, M. 2000. Computational aspects of kinetic analysis. Part B: The

ICTAC Kinetics Project Ð the decomposition kinetics of calcium carbonate revisited,

or some tips on survival in the kinetic minefield. Thermochimica Acta, 355, 145-154.

MENKART, J., WOLFRAM, L.J., MAO, I., 1984. Caucasian hair, Negro hair, and

wool: similarities and differences. Journal Society of Cosmetic Chemists, 17, 769-787.

MILCZAREK, P., ZIELINSKI, M., GARCIA, M. L., 1992. The mechanism and

stability of thermal transitions in hair keratin. Colloid Polymer Science, 270, 1106-

1115.

MONTEIRO, V. F., MACIEL, A. P., LONGO, E., 2005. Thermal Analysis of

Caucasian Human Hair. Journal of Thermal Analysis and Calorimetry 79, 289-293.

OPFERMANN, J. 2000. Kinetic Analysis Using Multivariate Non-linear Regression. I.

Basic concepts. Journal of Thermal Analysis and Calorimetry,, 60, 641-658.

OZAWA, T. 1970. Kinetic Analysis of Derivative Curves in Thermal Analysis.

Journal of Thermal Analysis and Calorimetry, 2, 301-324.


160

POPESCU, C., SEGAL, E., 1998. Critical considerations on the methods for

evaluating kinetic parameters from nonisothermal experiments. International Journal

of Chemical Kinetics, 30, 313-327.

POPESCU, C., WORTMANN, F. J., 2003. HPDSC evaluation of cosmetically treated

human hair. Roumanian J Chemistry, 48, 981-986.


Springer-Verlag.

RÜEGG, M., MOOR, URSULA., BLANC, B., 1977.

A calorimetric study of the thermal denaturation of whey proteins in simulated milk ul

trafiltrate. Journal of Dairy Research, 44, 509-520.

SAKABE, H., ITO, H., MIYAMOTO, T., INAGAKI, H., 1987. States of Water

Sorbed on Wool as Studied by Differential Scanning Calorimetry. Textile Research

Journal, 57, 66-72.

SANCHEZ-RUIZ, J. M. 1992. Theoretical analysis of Lumry-Eyring models in

differential scanning calorimetry. Biophys J, 61, 921-935.

SCHAWE, J. E. K., HOHNE, G. W. H., 1996. The analysis of temperature modulated

DSC measurements by means of the linear response theory. Thermochimica Acta, 287,

213–223.

SCHICK, C. 2002. Temperature modulated differential scanning calorimetry (TMDSC)

- basics and applications to polymers. In: CHENG, S. Z. D. (ed.) Handbook of

Thermal Analysis and Calorimetry.

SCHMIDT, H., WORTMANN, F. J., 1994. High pressure differential scanning

calorimetry and wet bundle tensile strength of weathered wool. Textile Research

Journal, 64, 690-695.

SCHWENKER, R. F., DUSENBURY, J. H., 1960. Differential Thermal Analysis of

Protein Fibers Textile Research Journal, 30, 800.

ŠESTAK, J., BERGGREN, G., 1971. Study of the Kinetics of the Mechanism of

SolidState Reactions at Increasing Temperatures. Thermochimica Acta, 3, 1-12.


161

SPARROW, L. G., DOWLING, L. M, LOKE, V. Y, STRIKE, P. M., 1988. Amino

acid sequence of wool keratin IF proteins. In: ROGERS, G. E., REIS, P. J, WARDS,

K. A, MARSHALL, R. C., (ed.) The Biology of Wool and Hair. London: Chapman &

Hall.

SPEI, M. 1984. Time-dependent deposition of anionic detergents in a-Keratins.

COLLOID & POLYMER SCIENCE, 262, 596.

SPEI, M. 1990. Thermoanalytical methods and their meaningfulness in keratin

research. Melliand Textilberichte, 11, 902.

SPEI, M., HOLZEM, R., 1989. The α-β transition of fiber keratins. Melliand Textilber,

70, 371.

SPEI, M., HOLZEN, R., 1987. Thermoanalytical investigations of extended and

annealed keratins. Colloid Polymer Science, 265, 965-970.

SPEI, M., THOMAS, H., 1983. Thermoanalytische Untersuchungen von

Wollproteinen in der Disulfidform. Colloid Polym. Sci., 261, 968-969.

STOUT, P. R., ROPERO-MILLER, J. D., BAYLOR, M. R., MITCHELL, J. M., 2007.

Morphological changes in human head hair subjected to various drug testing

decontamination strategies. Forensic Science International, 172, 164-170.

THIBAUT, S., BERNARD, B. A., 2005a. The biology of hair shape. Int. J. Dermatol,

44, 2-3.

THIBAUT, S., GAILARD, O., BOUHANNA, P., CANNELL, D. W., BERNARD,

B. A., 2005b. Human hair shaped is programmed from the bulb. British Journal of

Dermatology, 152, 632-638.

TONIN, C., ALUIGI, A., BIANCHETTO SONGIA, M., D'ARRIGO, C.,

MORMINO, M., VINEIS, C., 2004. Thermoanalytical Characterisation of Modified

Keratin Fibres. Journal of Thermal Analysis and Calorimetry, 77, 987-996.

VYAZOVKIN, S., SBIRRAZZUOLI, N., 2006. Isoconversional Kinetic Analysis of

Thermally Stimulated Processes in Polymers. Macromolecular Rapid Communications,

27, 1515-1532.


162

WATT, I. C. 1980. Sorption of Water Vapor by Keratin Journal of Macromolecular

Science, Part C: Polymer Reviews, 18, 169-245.

WATT, I. C., KENNETT, R. H., JAMES, J. F. P., 1959. The Dry Weight of Wool.

Textile Research Journal, 29, 975-981.

WEI, G. H., BHUSHAN, B., TORGERSON, P. M., 2005. Nanomechanical

characterization of human hair using nanoindentation and SEM. Ultramicroscopy, 105,

248-266.

WHITELEY, K. J., KAPLIN, I. J., 1977. The comparative arrangement of microfibrils

in ortho--, meso-, and para-cortical cells of Merino-wool fibers. J.Text.Inst, 68, 384-

386.

WORTMANN, F. J., DE JONG, S., 1985. Analysis of the humidity-time superposition

for wool fibers. Text Res J. , 55, 750-756.

WORTMANN, F. J., DEUTZ, H., 1993. Characterizing keratins using high-pressure

differential scanning calorimetry (HPDSC). Journal of Applied Polymer Science, 48,

137-150.

WORTMANN, F. J., DEUTZ, H., 1998. Thermal analysis of ortho- and para-cortical

cells isolated from wool fibres. Journal OF Applied Polymer Science, 68, 1991-1995.

WORTMANN, F. J., HULLMANN, A., POPESCU, C., 2007. Water management of

human hair. International Journal of Cosmetic Science, 30, 388-389.

WORTMANN, F. J., POPESCU, C., SENDELBACH, G., 2006. Nonisothermal

denaturation kinetics of human hair and the effects of oxidation. Biopolymers, 83, 630-

635.

WORTMANN, F. J., SPRINGOB, C., SENDELBACH, G., 2002. Investigations of

cosmetically treated human hair by differential scanning calorimetry in water. J.

Cosmet. Sci., 53, 219-228.

Chapter 6 FTIR Investigation of Human Hair

163

Chapter 6 FTIR Investigations of Human Hair

6.1 Introduction

Almost every compound, whether organic or inorganic, has covalent bonds. Different

bonds absorb different frequencies of electromagnetic radiation in the infrared region

of the electromagnetic spectrum. Although some of the frequencies absorbed in two

different molecules might be the same, no two molecular structures produce the

identical infrared spectrum. Therefore, infrared spectroscopy (IR) can result in a

qualitative analysis of each material. Furthermore infrared spectroscopy is also an

excellent tool for quantitative analysis, using specialised software tools. The size of the

peaks in the spectrum provides useful information of the amount of material, e.g., in a

mixture (Hsu, 1997).

In infrared spectroscopy, IR radiation is passed through a sample. Some of the infrared

radiation is absorbed by the sample and some of it is transmitted. The produced

absorption or transmission spectrum represents a characteristic fingerprint of the

sample. An infrared spectrum represents a fingerprint of a sample with absorption

peaks, which generally presents wavelengths or wavenumber as the x-axis and

absorption intensity or percent transmittance as the y-axis. Transmittance (T) is the

ratio of radiant power transmitted by the sample (I) to the radiant power incident on the

sample (I0). Absorbance (A) is the logarithm to the base 10 of the reciprocal of the

transmittance (T) (see Equation.6.1):

A= ln (1/T) = -lnT = -lnI/I0 (6.1)

The transmittance spectra provide better contrast between intensities of strong and

weak bands because transmittance ranges from 0 to 100% whereas absorbance ranges

from infinity to zero. It should be noted that the same sample can give different IR

spectra, such as one spectrum presented in wavenumber, while another is presented in


164

wavelength. Depending on this IR bands may become contracted or expanded (Khan,

2012).

The IR region is categorised into three smaller areas: Near IR, Mid IR and Far IR, as

shown in Table 6.1.

Table 6.1: Characteristics of the three IR regions

Near IR Mid IR Far IR

Wavenumber 13,000-4,000 cm-1

4,000-200 cm-1

200-10 cm-1

Wavelength 0.78-2.5 μm 2.5-50 μm 50-1000 μm

This FTIR investigation of human hair focus in the mid IR region, between 4000 and

400cm-1

(2.5 to 25 μm). The far IR region is used for analysis of organic, inorganic,

and organo-metallic compounds, which contain heavy atoms (mass number over 19). It

is useful in structural studies such as conformation and lattice dynamics of samples.

There is minimal or no sample preparation required for the near IR spectroscopy. It

provides high-speed quantitative analysis without consumption or destruction of the

sample, and is gaining increased attention in process control applications.

In the mid-infrared (MIR) spectrum, different functional chemical groups lead to

unique fundamental vibrations. An important use of the infrared spectrum is to

determine structural information about a molecule. Different atoms in the molecules

vibrate and rotate at particular wavelengths. The frequencies at which the molecules

absorb depend on the masses of the atoms, the force constants of the bonds, and the

geometries of the atoms. The absorption bands for different moieties are generally

known, so an unknown spectrum can be compared to the known one to identify the

material. In general, C-C, C-O, and C-N absorb between 1300-800 cm-1

, C=O, C=C,

C=N, and N=O absorb between 1900- 1500 cm-1

, C≡C and C≡N absorb between 2300-

2000 cm-1

, and C-H, O-H, and N-H absorb between 3800-2700 cm-1

(Silverstein, 1981).


165

6.1.1 Infrared Spectrometer

The instrument that determines the absorption spectrum for a compound is called an

infrared spectrometer or a spectrophotometer. Two types of infrared spectrometers are

in common use: dispersive and Fourier transform (FT) instruments. Both of these types

of instruments provide spectra of compounds in the common range of wavenumbers of

4000 to 400 cm-1

.

6.1.1.1 Dispersive Infrared Spectrometer

"Dispersive" or "scanning monochromator" is an alternative method for taking spectra,

which measures radiation intensity over a narrow range of wavelengths at one time. In

the dispersive system, infrared light is separated and scanned with the whole range of

infrared frequencies so the energy is sequentially presented to the detector. The

dispersive method is more common in UV-Vis spectroscopy, but is less practical in the

infrared than the FTIR method. A FT infrared spectrometer provides the infrared

spectrum much more rapidly than a dispersive instrument.

6.1.1.2 Fourier Transform Infrared Spectrometer

FTIR is the preferred method of infrared spectroscopy, which is used to obtain infrared

spectra of absorption, emission, photoconductivity of a solid, liquid or gas.

Fourier transform infrared (FTIR) spectrometers have been widely used in modern

infrared spectrometers and have improved the acquisition of infrared spectra. The

schematic diagram in Figure 6.1 represents the Michelson interferometer, which is the

central unit of a FTIR spectrometer. It has a fixed and a movable mirror. When

radiation from a broadband source strikes the beamsplitter, some of the light is

transmitted to a movable mirror and some of the light is reflected to a fixed mirror. The

moving mirror generates a variable optical path between two beams which gives

detector signal about the spectral information. A beam splitter splits the emitted light


166

from the light source: half of the initial light is reflected towards the fixed mirror and

then reflected back towards the beamsplitter, where about 50% passes to the detector.

The other half passes the beam splitter on its first encounter, is reflected by the

movable mirror back to the beamsplitter, where 50% of it is reflected towards the

detector (Barth, 2007). When the two beams recombine they interfere constructively or

destructively depending on the optical path difference to produce a pattern called an

interferogram (Pavia, 1996, Harris, 1999). The interferogram contains all the

frequencies, which form an IR spectrum. A mathematical operation known as a Fourier

transform is used to distinguish the individual absorption frequencies from the

interferogram to produce a plot of intensity versus frequency, i.e. an IR spectrum

(Pavia, 1996, Harris, 1999).

Figure 6.1: Schematic diagram of an FTIR spectrometer, showing its layout

(ThermoNicolet, 2001)


167

6.1.1.2.1 The Advantages of FTIR

FTIR instruments have several significant advantages over older dispersive

instruments (Barbara, 2004).

1. Multiplex advantage (Felgett advantage) – Many scans can be completed and

combined on an FT-IR in a shorter time, resulting in faster data collection of an

FT-IR spectrum and the improvement of the signal-to-noise ratio (SNR).

2. Throughput advantage (Jacquinot advantage) – The total source output can be

passed through the sample continuously, resulting in a substantial gain in

energy at the detector, hence translating to higher signals and improved SNRs.

3. Co-addition of scans – Increase SNR by signal-averaging, leading to an

increase of signal-to-noise proportional to the square root of the time, as

follows:

SNR α n½

4. High scan rate – The mirror has the ability to move over short distances rapidly

to acquire spectra on a millisecond timescale.

5. High resolution – By closing down the slits, a narrow band is achieved.

6. Laser Referencing (Connes Advantage) – By using a Helium-Neon laser as a

reference, the wavenumber calibration of interferometers is much more

accurate and has much better long term stability than the calibration of

dispersive instruments.

7. Negligible stray light – The detector responds only to modulated light, so there

is no direct equivalent of the stray light found in dispersive spectrometers.

8. Powerful computers – Advances in computers and new algorithms have

allowed for fast Fourier-transformation.

The main advantage of Fourier transform spectrometers is their rapid data collection

and high light intensity at the detector and in consequence a high signal to noise ratio.

Different types of detectors, light sources and other optical components are chosen to

optimise different requirements of the infrared spectral spectrum.

These advantages make FT-IR an invaluable tool for quality control and assurance

applications. It can detect even the smallest contaminants because of its sensitivity. In


168

addition, the accuracy of FT-IR detectors, as well as a wide variety of software

algorithms, have dramatically improved the applicability of infrared spectroscopy for

quantitative analysis, as it is non-destructive in nature.

6.1.2 FTIR Sampling Techniques

Infrared spectroscopy is a versatile analytical technique, which is feasible for liquids,

solids, and gases. However, many materials are opaque to IR radiation and must be

dissolved or diluted in a transparent matrix in order to obtain spectra, which is the

traditional transmission method of infrared spectroscopy (Hsu, 1997). Alternatively, it

is possible to obtain reflectance or emission spectra directly from opaque samples,

though attenuated total reflectance. A wide range of sampling accessories is employed

to take advantage of the capabilities of FTIR instruments.

6.1.2.1 Transmission spectroscopy measurements

Transmission spectroscopy is the most basic infrared method by transmitting the

infrared radiation directly through the sample at specific wavelengths, as shown in

Figure 6.2. However, sometimes sample preparation is required. When the sample is

liquid or solid, its intensity of the spectral appearance depends on the thickness of the

sample which is generally between 1 and 20 μm thick (Smith, 1996).

Sample

Figure 6.2: An illustration of transmission FTIR sampling, where the infrared beam

passes through the sample and then strikes the detector

Detector IR Beam


169

6.1.2.1.1 Sample Preparation

A. Liquids

In conventional transmission infrared spectroscopy, the IR beam passes directly

through the solution. To minimise effects of IR light scattering, samples should be

presented in thin films or suspensions.

In the process of sample preparation, a drop of a liquid organic compound is placed

between a pair of polished sodium chloride or potassium bromide plates, which is

referred as salt plates. A thin liquid film is produced by gently pressing the plates.

Since there is no solvent is involved, a background spectrum is considered as a neat

spectrum (Pavia, 2001).

Water is an intense absorber. One drawback of aqueous solutions is the strong

absorbance of water in the mid-infrared spectral region (near 1645 cm−1

) (Venyaminov,

1997), which masks or overlaps the important amide I band of proteins and some side

chain bands of interest. Thus protein solutions cannot be used for structure analysis of,

e.g., the α- and β- conformation. And organic compounds analysed by this technique

must be free of water.

B. Solids

There are three general methods for preparing solid samples in transmission

spectroscopy: alkali halide pellet, mulls and films.

I. Alkali Halide Pellet or Discs

This method involves mixing a solid sample (a few mg) with a dry alkali halide

powder (100-200 mg). The mixture is usually ground in an agate mortar with a pestle

under sufficiently high pressure in an evacuable die. The most commonly used alkali

halide is potassium bromide (KBr), which is completely transparent in the mid-infrared

region. But KBr is hygroscopic, so that the “water effect” may interfere with the


170

obtained spectrum. In addition, the preparation of a KBr pellet is very time consuming

and sample weight and grinding conditions may cause further problems in the infrared

spectrum of a sample (Painter, 1978, 1981). As light scattering shifts the spectral

baseline towards the high-frequency region (Painer, 1985), in order to minimize band

distortion, the sample should be ground to particles of 2 μm (the low end of the

radiation wavelength) or less in size (Hsu, 1997).

II. Mulls

This method is used as alternatives for pellets. It involves grinding the sample (no

more than 20 mg) and suspending it (about 50mg) in 1-2 drops of a mulling agent. The

mixture is further ground until a smooth paste is obtained, which is then spread

between two mid-infrared transparent windows e.g. NaCl, KBr, CaF2. Ideally, a

mulling agent should be infrared transparent between 4000 and 600 cm -1

,

unfortunately no such agent exists. The most commonly used mulling agent is Nujol

(liquid paraffin), but mineral oil obscures the bands that may be present in the analysed

compound. Normally, Nujol bands appear at 2924, 1462, and 1377 cm-1

(Pavia, 2001).

Similarity as for the alkali halide pellet, sample particles of 2 μm or less in size are

very important for obtaining a satisfactory spectrum.

III. Films

Films can be produced by either solvent casting or by melt casting. These methods are

particularly useful for examining polymers.

In solvent casting the sample is dissolved in an appropriate solvent, which is chosen

not only to dissolves the sample, but also to produce a uniform film. The concentration

of solvent depends on the required film thickness. The solution of the sample is poured

onto a glass plate (such as a microscope slide) or a metal plate and is spread to a

uniform thickness. Once the solvent has evaporated in an oven, the film is free to be


171

isolated from the plate. Alternatively, it is possible to cast a film directly onto the

infrared window.

Solid samples which melt at relatively low temperatures without decomposition can

also be prepared by melt casting. A film is achieved by hot-pressing the sample in a

hydraulic press between heated metal plates (Barbara, 2004).

6.1.2.1.2 Sample Preparations of Transmission Spectroscopy for Keratin Fibres

Various researchers have attempted various methods for preparing fibrous and elastic

materials for infra-red spectral investigations. Beer et al. (1959) cut fibrous proteins

into small pieces with a microtome, while Crook and Taylor (1958) mulled the textile

material between ground-glass plates with a few drops of paraffin or

hexachlorobutadiene. Devaney and Thompson (1958) used diamond powder embedded

in steel to grind a powder of epoxy-resin from a film of the epoxy resin and hardener.

There is some evidence using the microtome that it affects the conformational structure

of the fibre (Caroti, 1956), while the grinding procedure described above reduces the

crystallinity of the wool and human hair. This finding is also in agreement with the

observations made by other researchers (Beer, 1959). Strasheim et al. (1961) adopted

the Wig-L-Bug, which is a wrist-type mortar with a hard-tempered steel vial has been

found most suitable for the KBr disk technique.

But one of the main problems with the transmission IR spectroscopy is that the fibres

are impossible to arrange in a cross-sectional direction because the fibres are randomly

mixed with alkali halide pellet. If the proper sample preparation for the fibre cross-

section could be developed, IR spectroscopy could be used to examine specific

locations in the fibre’s internal structure.

6.1.2.2 Reflectance Techniques

Reflectance technique is an alternative method to analyse the samples, when they are

difficult to analyse by the transmittance methods. Reflectance methods can be divided

into two categories: internal reflectance measurement, which is using an attenuated


172

total reflectance cell in contact with the sample, and external reflectance measurement

which involves an infrared beam reflected directly from the sample surface.

6.1.2.2.1 Attenuated Total Reflectance Spectroscopy

Attenuated Total Reflectance (ATR) Spectroscopy, also known as Internal Reflection

Spectroscopy (IRS), has widely been used for the study of the near-surface of

biological materials, and has been used for investigating samples which are too thick

for transmission measurements (Harrick, 1979).

Generally, ATR spectra are similar to regular transmission spectra. But ATR sampling

does not produce totally absorbing spectral bands because its effective path-length is

controlled by the crystal properties, thereby the sample re-preparation time is reduced.

Moreover, ATR spectra have both absorption and reflection features, so they cannot be

used in quantitative analysis directly (Grdadolnik, 2002).

A. The principle of Attenuated Total Reflectance Spectroscopy

In FTIR-ATR, the sample is placed on a crystal with an index of refraction that is

larger than that of the sample, such as diamond, zinc selenide (ZnSe), ZnS, germanium

(Ge) and thallium/iodide (KRS-5). Different ATR cells properties are used for samples

in liquid or solid form. For a liquid sample, ATR produces a less intense solvent

contribution for the overall infrared spectrum, so that solvent spectra can be easily

subtracted from the sample spectrum of interest. Solid samples are best analysed by

single reflection ATR accessories, and diamond is widely used for most applications

because of its robustness and durability.

ATR spectroscopy examines the phenomenon of total internal reflection. A beam of

radiation is transmitted into a crystal and directed towards the sample interface at an

angle of incidence. When the incidence angle θi , at the interface between sample and

crystal exceeds the critical angle θc, total internal reflection is achieved. Total internal

reflection at the surface of the crystal produces an evanescent wave, which has the

same frequency as the incoming light. But the amplitude of the electric field decays


173

exponentially with the distance from the interface. The evanescent wave penetrates the

surface of the sample, and some of the light diffuses and is selectively absorbed in the

sample. When diffused light reflects back into the crystal, this part of the light carries

the information about the infrared spectrum of the sample. Afterwards it reaches the

detector to produce an IR spectrum. Because the IR beam is not transmitted through

the sample, the produced spectra are unaffected by turbidity. The attenuated radiation

is measured and plotted as a function of the wavelength to give the absorption spectral

characteristics of the sample.

B. ATR-diamond cell Investigation on Hair Fibres

FTIR spectroscopy using attenuated total reflectance (ATR) is an established technique

for non-destructive qualitative and quantitative surface analysis of keratin fibres,

namely, for cysteic acid and intermediate oxidation products of cystine. Researchers

employed an ATR diamond cell to measure the concentration of cysteic acid in

bleached hair. Although the intensity differences between untreated and bleached hair

were not discussed, the method of data analysis as described by Signori (1997) is

excellent for the determination of the cysteic acid content of both damaged and

untreated hair. Martin (1999) further compared ATR using ZnS cells and diamond

cells, confirming that the diamond cell ATR has good applicability for hair keratin

analysis because of its reproducible results. Although the ATR technique is not

suitable for single hair fibre measurements, it will provide information about the

surface of the hair sample (up to 4μm into the bulk of the fibre) (Signori, 1997).

ATR is a powerful IR method as it is insensitive to sample thickness, less destructive

(only a small point of the fibre is compressed by the pressure tower), and permitting

the surface or near-surface analysis of thick or highly absorbing materials, i.e. α-

keratin fibres (Bartick, 1994, Signori, 1997). But IR radiation does not penetrate the

hair fibre, so a lot of reflectance may occur on the hair fibres surface. Thus ATR can

only investigate the cuticle and possibly outer layers of the cortex.


174

C. Depth of Penetration in ATR

The depth of penetration (DP) in an ATR experiment is a measure of how far the

evanescent wave penetrates into a sample. Equation 6.2 for depth of penetration in an

ATR experiment yields:

DP = λ/ 2π n1 {sin2θ – (n2/n1)

2}

1/2 (6.2)

Where,

DP = Depth of Penetration (in cm)

λ = Wavelength of the infrared light in cm-1

n1 = Refractive index of ATR crystal

θ = Angle of incidence of IR beam with crystal surface

n2 = refractive index of sample

Generally the depth of penetration in ATR ranges from about 0.1 μm up to about 5 μm

depending on the refractive index of the crystal and the sample (Smith, 1996), the

frequency and the incident angle of the IR beam (Harrick, 1985). The penetration

depth, dp, is defined as the distance at which the amplitude of the evanescent wave has

decreased to (1/e) or 37% of its original value at the surface (Smith, 1996). Figure 6.3

indicates that the strength of the evanescent wave decreases exponentially in intensity

from the surface of the crystal.

Figure 6.3: Schematic diagram of a single reflection ATR showing the radiation path

through an infra-red transmitting crystal of high refractive index. The sampling area of

the crystal makes contact with the sample. IR radiation from the spectrometer is

deflected firstly via a mirror to the ATR crystal, in which it then undergoes total

internal reflection, penetrating superficially into the sample, before exiting the crystal

and being re-directed to the detector.

θ

ATR Crystal

Bulk Sample

Evanescent Wave

Dp


175

6.1.3 FTIR Technique and Protein Structure

6.1.3.1 Introduction

The FTIR technique measures the absorption peaks of various infrared wavelengths

and, thus provides information about the chemical composition of various polymers,

including amino acids from proteins. FTIR has been widely used in the research of

wool and other hair fibres, which are natural protein fibres.

It has been suggested that at the molecular level, FTIR analysis can provide rapid and

specific structural information of proteins and their composition (Panayiotou, 1999).

Therefore, a brief introduction to protein structure is given before discussing the

molecular structure in the sample. Proteins are linear biological polymers, for which

the monomeric units are amino acids. In general, protein may have about twenty

different amino acids, which are distinguished by the identity of the “R” group (side

chains of different structure). The amino acids are bonded together to form a polymer

by linking the amino group between one amino acid with the carboxylic acid group

with adjacent amino acid to form an amide bond (Figure 6.4).

R

CC

H

OHN

H

H

O

Figure 6.4: The structure of an amino acid: the basic building block for making

proteins

In all proteins, the amide bond is the most abundant structure. The atoms take part in

this bond lead to a number of vibrational bands that can be observed in the IR

spectrum of proteins. Characteristic bands found in the infrared spectra of proteins and

polypeptides include the Amide I, Amide II, and III bands, which absorb in the 1600-

1700, 1500-1600, and 1200-1350 cm-1

regions, respectively. These regions arise from

the amide bonds which are used to link the amino acids. The absorption associated

Carboxylic acid group

Alpha carbon

“R” group

Amino group


176

with the Amide I band is predominantly (80%) attributed to the C=O stretching mode

of the amide functional group, while the remaining contribution (20%) arises from C-

N stretching. The absorption associated with the Amide II band primarily originates

from the bending vibrations of the N—H bond and C-N stretching vibrations. The

amide III absorption is caused by coupling C-N stretching vibrations with N-H in

plane bending vibrations, and the weak contributions from C-C and C=O stretching

(Figure 6.5).

C

O

NH H

H

C N

O

HH

H

Figure 6.5: The vibrations mainly responsible for the Amide I and Amide II bands in

the infrared spectra of proteins and polypeptides. The Amide I band is due to carbonyl

stretching vibrations while the Amide II is due to NH bending vibrations.

Other vibration modes also appear in the Mid-IR range, include the amino acid side

chains, such as C-H deformations (1471-1460 cm-1

), CH2 and CH3 deformations (1453-

1443 cm-1

and 1411-1399 cm-1

), and the cystine oxide stretches, such as the

asymmetric and symmetric cysteic acid (1171 cm-1

and 1040 cm-1

), symmetric cystine

dioxide (1121 cm-1

), and cystine monoxide (1071 cm-1

) stretches.

6.1.3.2 FTIR Peak Assignments on Human Hair

6.1.3.2.1 Introduction

In FTIR-ATR spectroscopy, the beam of radiation penetrates the outer surface of the

hair, which is the cuticle. The cuticle has a high concentration of cystine, along with

other amino acids which are not always found in α-helical polypeptides. Oxidising

agents such as alkaline hydrogen peroxide can break the disulfide bond and oxidise the

sulphur atoms to sulphonate groups. Other oxidation products involve the addition of

one or more oxygen atoms to the disulfide group, so that sulphoxide or sulfone

linkages are also formed. These intermediates are not very stable. Once the hair keratin

Amide I vibration Amide II vibration


177

is subject to hydrolysis, they are converted to cysteic acid. Since sulphonate,

sulphoxide, and sulphone groups can be distinguished from one another by their infra-

red absorption (Bellamy, 1958), with the application of FTIR spectroscopy the S-O

vibrations can be analysed and cystine monoxide and cystine dioxide can be detected.

Figure 6.6 summarises the structures of the most important sulphur-containing amino

acids in bleached hair.

Cysteine Cystine

CH2 CH COOH

NH2

SH

S S CH2CH2CHHOOC

NH2

CH2

NH2

COOH

Cysteic acid Cysteine-S-thiosulphate

CH CH2 SO3H

NH2

HOOC

CH CH2HOOC

NH2

S SO3H

Figure 6.6: The structures of the most important sulphur-containing amino acids in

bleached hair

6.1.3.2.2 Literature Review

Methods capable of measuring oxidative hair damage are of great interest to the hair

care industry. FTIR spectroscopy had been applied to study the effects of oxidative

treatment on human hair fibres for a long time (Robbins, 1967, Strasheim, 1961, Shah,

1968, Alter, 1969). Brenner et al. (1985) investigated untreated and bleached hair

fibres by using FTIR spectroscopy with a diamond anvil cell. For the bleached hair

fibres, they discovered the presence of a peak at 1044 cm-1

, which was attributed to the

symmetric stretch of cysteic acid. This peak was considered to discriminate between

treated and untreated hair samples. Ohnishi et al. (1993) furthered this study by

analysing permanently waved hair fibres. In his investigation, the concentration of

cysteic acid and random damage patterns increased along the hair length.

Alter et al (1969) mention that new peaks appear at 1175 and 1040 cm-1 for bleached

hair (treated with H2O2) for various oxidising times. These peaks correspond to

sulphonates and cysteic acid, respectively. The assignment of the bands as sulfonate

linkages from cysteic acid residues seems well established (Colthup, 1964, Bellamy,

1975). Cysteic acid residues arise from disulfide bond fission of cystine, which is the

most abundant amino acid in hair (Robbins, 1994). Cystine dioxide peaks (1220 and


178

1310 cm-1

) and increased sulphonates intensity are also observed at 1175 cm-1

from

oxidation of cysteine. These peaks show an increase in intensity after successive

bleaching treatment, resulting in the overall shape of the bleached hair spectrum is

significantly different from the untreated hair spectrum.

Kirschenbaum et al (2000) examined some structural information changes for treated

hair samples. Their results showed not only the disulphide bond breakage, but the

peptide backbone was also affected by chemical treatments. The damaged hair

spectrum indicated an increased intensities of C-H peaks, which was attributed to the

structural transformation of α−helices to β−sheets.

Previous FT-IR spectroscopic studies have assigned several absorbance frequencies to

particular component amino acids (Carr, 1993, Lewis, 1993, Fredline, 1997, Pielesz,

2000). It was shown that hair samples including the untreated sample show cystine

derivatives. IR peaks at 1040, 1075, 1175, and 1229 cm-1

correspond to cystine

oxidation products, but the other peaks at 1241, 1454, 1550, and 1655 cm-1

correspond

to the amino acids. Table 6.2 shows the peak assignments of cystine derivatives and

amino acids.

Table 6.2: The Peak Assignments for Cystine Derivatives and Amino Acids in the

mid-IR Spectrum

Wavenumbers (cm-1

) Assignment

1040 Sulphonate, S-O sym, stretch

1075 Cystine monoxide (R-SO-S-R)

1175 Sulphonate, S-O asym, stretch

1229 Cystine dioxide (R-SO2-S-R)

1241 Amide III (N-H stretch)

1454 CH2

1550 AmideII (C-N stretch)

1655 Amide I (C=O stretch)

2885 CH

2960 CH

3315 NH (Primary amine)

3550 OH (H2O)


179

6.1.3.3 FTIR Investigation on Human Hair

Cystine is an important amino acid in hair keratin fibres, as its disulphide bond plays

an important role in determining the mechanical and physical properties of the fibre. It

has been confirmed that IR spectroscopy can detect the oxidation products of cystine,

i.e. cystine monoxide, cystine dioxide, and cysteic acid (Hilterhaus-Bong, 1987),

without directly supplying spatially resolved information.

Other FTIR studies on hair include analysing the hydrogen bonds around hair protein

to understand the mechanism of improving the durability of hair-setting with acid and

ethanol treatment (Itou, 2006) and also studying isolated melanins from fair, red and

dark hair (Bilinska, 1991). IR has also been used to study other keratin fibre, such as

horse hair (Lyman, 2001), plant cutin (Benítez, 2004) and human skin, nails, and lipids

(Gniadecka, 1998).

6.2 Objectives

The IR- spectrum of bleached hair is different from untreated hair. Since cysteic acid is

the most abundant amino acid in hair, its oxidation can be studied by FTIR

spectroscopy. Reflectance and transmission are two modes of IR spectroscopic

measurements. Reflectance mode measurements are easier to obtain; however,

variations in the surface properties of the sample may influence calibrations.

Transmission mode may be less susceptible to surface properties, but the amount of

light penetrating the sample is a problem (Schaare, 2000). The objective of this

research was to combine attenuated reflectance and transmission modes of FTIR

spectroscopy for detecting oxidation effect of cysteic acid caused by three types of

bleaching treatments. An ATR diamond cell accessory and KBr pellets were employed

in this investigation.


180

6.3 Experimental Work

The experimental work for the FTIR investigation includes sample preparations,

spectral processing and data analysis.

6.3.1 Sample Preparation

All hair samples were randomly collected, and the virgin hairs are considered as those,

which had never been subjected to chemical damage.

6.3.1.1 ATR Sample Preparation

As describe in Section 6.1.2.2.1, FTIR-ATR requires no sample preparation. The hair

fibre spectra were collected and recorded on a Thermo Nicolet 5700 FTIR

Spectrometer for an angle of incidence of 45° with a Diamond-ATR Smart Orbit

Accessory to provide information about the surface of the hair fibres in a spectral

range between 4000 - 400 cm-1

. The background spectrum was obtained for air (the

crystal in contact with no sample) before collecting the sample spectrum. 64 sample

scans were added at a resolution of 4 cm-1

to produce infrared spectrum for each

sample. Since hair is a heterogeneous material, its physical and morphological

properties may vary greatly, even thought hair may be collected from the same head.

As a result, the area from the tip at 2cm, 5cm, 8cm, 11cm and 14cm (commercial

bleaching only) was twisted tightly and measured on the same hair tress to examine the

uniformity of bleaching effects. Three tresses of each treatment were measured to

yield reproducible oxidation values. All spectrum analyses were carried out using the

OMNIC software (Nicolet).

In the spectral sampling process, the solid material is placed on the cleaned diamond

crystal surface after the background spectrum is collected, and then sufficient

consistent pressure is applied using the pressure tower to record the sample spectrum.

Any inadequate contact between sample and pressure tower could cause a narrow


181

transmission range, noisy spectra and poor reproducibility. Once the sample spectrum

had been recorded, the spectrum information is collected using OMNIC 7.2 spectral

software program (as .SPA files).

A. Calculation of the penetration depth

Hair diameter is a variable parameter that affects the result of FTIR-ATR. Oxidative

attack is likely to be concentrated in the cuticle and outer cortical cells of the human

hair. This is because of the impervious nature of the cuticle scales and of cystine being

more concentrated in the cuticle than in the cortex. The penetration depth (dp) of the

infrared beam into the hair sample was calculated as:

DP = λ/ 2π n1 {sin2θ – (n2/n1)

2}

1/2 (6.2)

The refractive index of hair (ηhair) is reported to be 1.555 (Baddiel, 1968). The depth of

penetration and the total number of reflections along the crystal can be controlled

either by varying the angle of incidence or by selection of crystals. Different crystals

have different refractive indices. When the Diamond/ZnSe crystal (n1=2.4) is used, the

FTIR-ATR penetration depth for a human hair fibre with a refractive index of 1.5 at

1000cm-1

(cysteic acid band, 1042cm-1

), is approximately 2.0µm when the incidence

angle is 45°. As described in Chapter 1, each cuticle cell is approximately 0.5μm thick

and the overall cuticle layer thickness varies between 5-10 layers (Robbins, 2002).

Hence, the average thickness of the cuticle layer for individuals is believed to be

around 3μm, which is much thicker than the penetration depth of hair (2μm) at

1000cm-1

. Thus the IR beam is not able to penetrate into the cortex, and FTIR-ATR

mainly focuses on the cuticle (surface) part of the hair fibres.

6.3.1.2 KBr Pellet preparation

Three batches of hair samples from each treatment were mixed and cut to a length of

approximately 2mm snippets, dried at 105oC for one hour to remove the moisture. A

bottle of sealed KBr powder was left in a 90o oven overnight before pellet preparation.

After removal from the oven, 200 mg dried KBr was immediately weighed and ground


182

to a fine powder. About 6-7 mg of hair snippets was added and mixed well with KBr

using a mortar and pestle by hand. This mixture was then transferred to a Graseby

Specac die that has a barrel diameter of 15 mm. Using a Carver press at a pressure of

10 kbar under vacuum for 2-3min a clear glassy disk was formed of about 4.5 mm

thickness (shown in Figure 6.7). The pellet is placed in a transmission holder to record

the spectrum.

Figure 6.7: KBr pellets of virgin and bleached hair and a pure KBr crystal

FT-IR spectra of hair samples, which are in a pellet form, were acquired with 5700 FT-

IR spectrometer of Thermo Nicolet equipped with KBr optics and deuterated triglycine

sulphate (DTGS) detector. The FT-IR data analysis was conducted using the infrared

spectra analysis software package OMNIC. All recorded spectra were average over 32

scans in the mid Infrared (IR) region of 4000-400 cm−1

by optimising the resolution of

the instrument at 4 cm−1

. A background spectrum was obtained as a reference through

a KBr pellet without fibres. The sample spectrum was obtained for the pellet

containing fibre snippets. Each time before obtaining the spectrum of a sample, a fresh

background spectrum from a KBr pellet was recorded. All the spectra were recorded

four times at different points to obtain reproducible results. The transmittance spectra

were then converted to absorbance spectra for further investigation.

6.3.2 Spectral Processing

In order to study the influence of different acquisition techniques (transmission and

attenuated total reflectance) for the quantification of oxidative hair damage incurred by

bleaching, different mathematical procedures (normalisation, second derivatives) are

applied to enhance peak resolution and to reduce the signal-to-noise ratio.

Virgin Hair

Pellet

Pure KBr Pellet 6%H202 Hair

Pellet

Commercial Persulphate Hair


183

6.3.2.1 Normalisation

All of OMNIC spectral (.SPA) files from ATR and Transmission measurements (KBr)

were imported and save as a .CSV file before transferring the data to a Microsoft Excel

2007 spreadsheet. In order to facilitate spectral comparison, normalisations of keratin

hair spectra were carried out based on the δ(CH2) stretching band (ca. 1452cm-1

). The

reason for choosing this wavelength as an internal standard is that this molecular bond

is associated with the amino acid side chains thus is not affected by cosmetic chemical

treatment, which otherwise results in peptide backbone conformational changes, e.g.

peroxides or thioglycolic acid or natural weathering processes, such as water or carbon

dioxide (Church, 1997, Kuzuhara, 2005). It has been confirmed that the intensity of

this band is consistent from sample to sample (Church, 1997).

6.3.2.2 Derivative Spectroscopy

Since Fourier transform infrared spectrum is a mathematical function, its derivative

can be calculated (Smith, 1996). The derivative spectroscopy describes the rate of

change of a signal recorded as a function of the wavelength or frequency (Bridge,

1987). For a given absorbance curve, the first derivative (dA/dλ) is the gradient of the

original spectrum at each wavelength. The first derivative of a spectrum is called a first

order derivatives, the derivative of the first derivative is called a second order

derivative and so on. The simplest way to calculate a spectral derivative is the “point

difference” method, where the difference in Y values between successive data points is

calculated, then plotted versus wavenumber (Smith, 1996). There are other ways of

calculating derivatives, depending on the software package used.

In a raw spectrum, the individual vibration peaks are difficult to identity as a

consequence of band broadening, thus, derivative spectroscopy is used to resolve

overlapping absorption peaks.

It can be seen that in the derivative spectroscopy the odd number derivatives show a

shift in the wavenumber of the peak, and the even numbered derivatives show the main


184

peak at the original wavelength of maximum absorption (Joy, 1991). Therefore it is

more practical to use the even numbered derivatives to interpret the spectra. However,

due to the spectra become increasingly complicated as the derivative order increases

(Joy, 1991), it becomes feasible to utilize the second order derivative to enhance the

resolution of spectrum for the further deconvolution or curve fitting analysis, since the

second order derivative method can at high resolution acquire the maximum

information from each spectrum, while minimising the background noise.

For the derivative analysis of FTIR spectra, the first step is to choose a part of the

spectrum containing bands of interest. The raw spectra (after normalisation) between

980 -1140 cm-1

was imported into Peakfit peak separation and analysis software

(version 1.2), and were converted into second derivatives spectra by using the

Gaussian deconvolution derivative method. As the PeakFit software decides the peak

position based on the second derivative, and peak fitting is followed for the closest

match to the original spectra, the data obtained by PeakFit seem to correspond reliably

to each real peak. The second derivative spectra were reduced to minimum signal-to-

noise ratio spectra using an automatic smoothing function, and the achieved spectra

were then imported to a Microsoft Excel 2007 spreadsheet for further investigation.

Also, the possibility of deconvolution of IR spectra was explored using second

derivative and PeakFit deconvolution software. This method improved peak selection

and peak fitting, however, the results depend on the values of the deconvolution

parameters, and it requires appropriate parameter settings.

6.4 Result and Discussion

6.4.1 FTIR spectrum of Human Hair

The vibrational spectrum of human hair keratin, particularly between 1750-900 cm-1

, is

very complex. This is due to the protein-polypeptide structure in hair keratin. Three

molecular structures that each generates characteristic vibrational absorption bands can

be identified within this fingerprint region:


185

(a) The peptide bond (primary protein structure). Caused by a condensation reaction

between the carboxylic acid and amine group of adjacent amino acids. It is the

most fundamental element within the keratin protein and yields the Amide I, II,

and III IR spectral bands.

(b) The secondary protein structure. Built up around the C-C skeletal stretching

backbone of all keratin proteins, it can exhibit one to three sensitive structure

conformations, those are the α-helical, β-sheet and random coil or amorphous

structures which are directly related to the Amide region.

(c) The amino acid side chains (R groups). The C-H vibrations originating from the

─CH, ─CH2 and ─CH3 of the aliphatic and aromatic rings of phenylalanine,

tyrosine, tryptophan and significant vibrations of the oxidative intermediates

from the amino acid cystine (i.e. S=O, SO2, SO3–, and –S–SO3

–)

The frequencies and peak assignments for virgin human hair in ATR mode and in

transmission are compared in Figures 6.8 and 6.9. They show that two hair spectra

exhibit many similarities in their characteristic bands, such as amide I, II, III, however

some variations in both the band positions and absorbance intensities and some

impurities in the transmission spectrum are observed. The impurities are represented

by the noise on the curve in the amide I region in Figure 6.8. They are thought to be

associated with water, whose vibration band is near 1650 cm-1

. Because of the

hydrophilic nature of KBr, moisture in the air is easily absorbed by the KBr during

sample preparation. As a result, the 1240 to 1750 cm-1

region of the spectrum is

excluded for further analysis. However, in order to make the spectrum of FTIR-ATR

and transmission mode comparable, a normalization of spectrum has to be applied

because of the systematic differences of the spectra.


186

Figure 6.8: The raw FTIR-ATR spectrum of an untreated hair sample. Spectral

assignments of some bands are made. The S-O vibration (represented as a solid bar on

the graph) is of particular interest in this research.

Figure 6.9: The raw FTIR-KBr spectrum of an untreated hair sample. Some bands are

assigned. The S-O vibration (represented as a solid bar on the graph) is of particular

interest in this research.

6.4.1.1 Assignment of Absorption Bands

As discussed in Section 6.4.1, the raw spectrum of α-keratin between 1750-900 cm-1

is

extremely complex, as there are many overlapping and broad bands. The application of

second derivative spectra resolves the complexity of keratin raw spectrum in Figures

6.10 and 6.11. However some peaks in a second derivative spectrum are too subtle to

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

750 950 1150 1350 1550 1750

Ab

sorb

ance

(A

)

Wavelength (cm-1)

Amide III

region Amide II region

S-O vibration

region

0.18

0.2

0.22

0.24

0.26

0.28

0.3

0.32

750 950 1150 1350 1550 1750

Ab

sorb

ance

(A

)

Wavelength (cm-1)

S-O vibration

region Amide II

region

Amide III

region

Amide I

region and

Water

Amide I

region


187

be detected. A peak in the second derivative located reasonably near to a known peak

wavenumber was considered as being a representative peak of the IR spectrum.

IR absorptions associated with cystine oxidation occur between 1200─1000 cm-1

. It is

obvious that both resolution-enhanced spectra in Figures 6.10 and 6.11 represent well

resolved peaks. Previous band assignments identified the symmetric and asymmetric

sulfonate S =O stretching vibrations at 1042 cm-1

and 1175 cm-1

(Weston, 1955,

Strasheim, 1961, Robbins, 1967, Alter, 1969), while we found these vibrations at 1039

cm-1

and 1171cm-1

in Figure 6.10 and at 1038 cm-1

and 1174 cm-1

in Figure 6.11.

The band at approximately 1039 cm-1

has been assigned to a symmetric S=O stretching

vibration of sulfonate or sulfonic acid groups of cysteic acid residues, and therefore its

absorbance is a measure of the – SO3- content of hair. In addition, another weaker anti-

symmetric cysteic acid band is shown at 1173 cm-1

, which is actually a shoulder of the

Amide III band. Since cystic acid is the most abundant amino acid found in hair,

followed by monoxide (1076 cm-1

), the absorbance peak heights of cysteic acid (1039

cm-1

) were measured and compared in order to measure the chemical changes arising

from bleaching of hair.

Other oxidative intermediate products also appear as oxidation bands in this region,

such as cystine monoxide (S-S=O) at 1076 cm-1

, cystine dioxide (S=O2) at 1123 cm-1

and cysteine-S-thiosulphate (Brunte Salt) at 1018 cm-1

. Previous researchers reported

bands near 1190, 1022 cm-1

for methyl and ethylthiosulfate ions, which they assigned

to the S=O asymmetric and symmetric stretch vibrations of the thiosulfate ions,

respectively (Simon, 1961).

In addition, the differences between ATR and transmission spectra in spectral

appearance and intensity of peaks depend on their sampling properties. Since

transmission analyses the internal region of hair, while FTIR-ATR samples the near-

surface chemistry only. Thus, with the application of these two methods, a

comprehensive result for oxidation of cysteine from the internal regions to the surface

of hair and for the three types of bleaching treatments is achieved.


188

Figure 6.10: Second derivative analysis of untreated hair (Batch I, 8cm) in the ATR


, which resolves the overlapping cysteic acid,

cysteine-S-sulphonate and cysteine monoxide peaks. Note the inversion of the peak

maxima with respect to the normal spectra (original data in absorption mode).

Figure 6.11: Second derivative analysis of untreated hair (Batch I) in the transmission


, which resolves the overlapping cysteic acid,

cysteine-S-sulphonate and cysteine monoxide peaks. Note the inversion of the peak

maxima with respect to the normal spectra (original data in absorption mode).

-1.50E-03

-1.00E-03

-5.00E-04

-2.00E-17

5.00E-04

980 1030 1080 1130 1180 1230

Arb

itra

ry U

nit

s

Wavelength (cm-1)

S-Sulphonate

1205 cm-1

Cysteic acid

1171 cm-1

Sulfenic acid

1153 cm-1

Monoxide

1076 cm-1 Cysteic acid 1039

cm-1

S-sulphonate

1016 cm-1

Monoxide

1065 cm-1

Inorg.Sulphate

1097 cm-1

-5.50E-04

-4.50E-04

-3.50E-04

-2.50E-04

-1.50E-04

-5.00E-05

5.00E-05

1.50E-04

980 1030 1080 1130 1180 1230

Arb

itra

ry U

nit

s

Wavelength (cm-1)

Dioxide

1123cm-1

Inorg.Sulphate

1099 cm-1

Monoxide

1076 cm-1

Cysteic acid

1038cm-1

S-Sulphonate

1018cm-1

Monoxide

1065cm-1

Dioxide

1123 cm-1

S-Sulphonate

1194 cm-1

Cysteic

acid 1174

cm-1

Sulfenic acid

1151cm-1


189

6.4.2 Comparison of FTIR Spectra of Untreated and Bleached

Human Hair

The cuticle cells, which protect the inner layers of hair, have a larger content of cystine

than the cortex (Robinson, 1976, Robbins, 2002), thus oxidative attack to the hair fibre

is likely to occur firstly in the cuticle and there mainly to the exocuticle, which is very

rich in cystine. For this reason, FTIR-ATR was firstly employed to investigate the

oxidation damage of the surface or near-surface regions of human hair.

6.4.2.1 ATR Technique

It has been discussed that in order to allow comparison of changes in cysteic acid peak

intensity after oxidation, all raw spectra have to be normalized against the peak

intensity of the C-H stretch band at 1452 cm-1

. The maximum oxidation duration (4h)

was taken as an example to illustrate the impact of oxidation on the cysteic acid

yielded by various bleaching conditions.

Figure 6.12 shows the relative peak heights of ATR–FTIR spectra that were

standardized against the peak heights at 1452 cm-1

. The differences in FTIR–spectra

between untreated and bleached hairs for different agents are shown in the range of

1300 - 900 cm-1

, where the cystine oxidation bands are located. The intense symmetric

cysteic acid band is located at 1040 cm-1

. A similar phenomenon is also observed with

the weaker anti-symmetric cysteic acid band at 1172 cm-1

. Commercial persulphate

bleach (4h) has the strongest effect on both cysteic acid bands, suggesting that the

commercial persulphate bleach has an extensive oxidation influence, followed by 9%

H2O2 commercial bleach, 6% H2O2 and then virgin hair.


190

Figure 6.12: FTIR-ATR spectra of untreated and 3 types of 4h treated fibers at 1300 –

900cm-1

, after normalization by against the peak intensity of the C-H stretch at 1452

cm-1

. The cysteic acid peak is clearly visible at 1040cm-1

and 1172 cm-1

for bleached

hair, (marked by the red arrows). Each curve is randomly chosen from one of three

tresses’ results.

The ATR second derivative spectra of three types of bleached hairs are compared in

Figure 6.13 with untreated hair to demonstrate the differences in cysteic acid formation

after 4h bleaching treatment. In principle, the second derivative contains three features

corresponding to each absorbance peak in the original spectrum, two pointing upward

and one pointing down. The valley in the second derivative corresponds to the

wavenumber of maximum absorbance of a band in the original spectrum. Since the

valley depth of the second derivative spectrum is a negative value, the absolute valley

depth is employed in the following discussion for the better understanding. The cysteic

acid peak is clearly visible as a valley at 1040cm-1

for the bleached fibre, confirming

the oxidation of disulphide.

Commercially persulphate bleached hair (4h) exhibits the most intensive valley depth

of cysteic acid, due to the aggressive oxidizing agent, persulfate, followed by 9% H2O2

commercial bleached (4h) hair, 6% H2O2 bleached (4h) hair and virgin hair. The

differences between the valley depths are associated with the fission of amino acid

chains by different damage extents of oxidation treatments, causing different amount

of cystine to be oxidised to cysteic acid.

0.6

0.8

1

1.2

1.4

900 950 1000 1050 1100 1150 1200 1250 1300

Ab

sorb

ance

aft

er

no

rmli

zati

on

Wavelength (cm-1)

Untreated hair 6% H202 Bleach 4h 9% H202 Commercial Bleach 4h Commercial Persulphate Bleach 4h


191

However, a slight shift in the valley position for cysteic acid after second derivative is

observed for the bleached hair. The valley moves from the expected 1042 to 1037cm-1

.

This is attributed to a change in disulphide conformation as a result of partial

disulphide bond rupture (Hogg, 1994).

Figure 6.13: FTIR-ATR spectrum 2nd

derivative of untreated and 3 types of 4h treated

fibers at 1300 – 900cm-1

. Cysteic acid becomes clearly visible as a valley at 1040cm-1

for the bleached fiber, (marked by the red arrow). Each curve is randomly chosen from

one of three tresses’ results.

6.4.2.1.1 Oxidation Effects along the Fibre Length

Due to the heterogeneity of hair fibre, different parts of hair are subject to damage

differently. In addition, hair also undergoes natural oxidative degradation processes

(weathering) due to its exposure to sun and atmospheric oxygen. To explore further the

variability of the hair samples after the three types of bleaches, five measurement

points from the root end of a tress at 2cm, 5cm, 8cm, 11cm and 14cm (commercial

bleaching only) are chosen. The corresponding results for the absolute valley depth for

cysteic acid in the second derivative spectra are presented in Tables 6.2 – 6.10.

Tables 6.2 – 6.10 demonstrate the changes in terms of intensity of cysteic acid absolute

valley depth (second derivative) through the extensive treatment time for bleached

hairs. All the bleached hairs show the progressively increased absolute valley depths

-0.007

-0.006

-0.005

-0.004

-0.003

-0.002

-0.001

0

0.001

0.002

0.003

980 1030 1080 1130

Arb

itra

ry U

nit

es

of

cyst

eic

aci

d p

eak

valle

y d

epth

Aft

er

Seco

nd

De

riva

tive

Wavelength (cm-1)

Untreated Hair

6% H202 Bleach 4h

9% H202 Commercial Bleach 4h Commercial Persulphate Bleach 4h


192

for cysteic acid. It is obvious that with longer bleaching time, higher intensities of

cysteic acid are observed. Taking a 5cm measurement from the root end, as an

example, the intensity of 6% H2O2 bleached hair for cysteic acid had a gradual increase

with the bleaching time, from 1.34 at 0.5h to 2.34 at 2h, and to 2.7 at 3h, and to 2.89 at

4h. Furthermore, 9% H2O2 commercial and commercial persulphate bleached hair

increased from 1.97 and 2.32 at 0.5h, to 5.45 and 4.62 at 2h, and to 5.39 and 5.56 at 3h,

and to 5.61 and 5.81 at 4h, respectively.

In addition, the higher peroxide concentration and the stronger oxidizing agent, the

higher the intensities of cysteic acid will be. The absolute valley depth for cysteic acid

intensity at the tip part (11cm) increases from 1.37 to 2.28, when comparing 6% H2O2

bleach with persulphate bleach at 0.5h in Table 6.3. Similarly 4h commercial

persulphate bleached hairs yielded a bigger absolute valley depth of cysteic acid 5.66

at the tip part (11cm), while the 6% H2O2 bleach has a cysteic acid intensity of 2.64 in

Table 6.10. These results are consistent with the findings that the tip of hair is more

damaged than the root end (Wolfram, 1971, Robinson, 1976, Tolgyesi, 1983), hence

they are more susceptible to oxidative attack.

Table 6.2: Absolute valley depth for cysteic acid of five measurement points of virgin

hair for the second derivative

Hair Types 2cm 5cm 8cm 11cm 14cm

Batch I 0.72 0.8 0.76 0.87

Batch II 0.8 0.73 0.94 1.02 1.05


bleached hairs for the second derivative at the bleaching time of 0.5h

2cm 5cm 8cm 11cm 14cm

6% H202 1.55 1.34 1.22 1.37

9% H2O2 Commercial 1.97 1.97 2.1 2.08

Commercial Persulphate 1.98 2.32 2.46 2.28 2.53


193


bleached hairs for the second derivative at the bleaching time of 1h


6% H202 1.88 1.89 1.69 1.91

9% H2O2 Commercial 2.93 2.99 3.14 3.41





6% H202 2.08 2.02 2.02 2.22

9% H2O2 Commercial 4.35 4.53 4.57 4.77





6% H202 2.18 2.34 2.39 2.43

9% H2O2 Commercial 5.34 5.45 5.39 5





6% H202 2.48 2.4 2.39 2.36

9% H2O2 Commercial 5.23 5.14 5.08 5.16



194




6% H202 2.7 2.7 2.41 2.44

9% H2O2 Commercial 5.66 5.39 5.21 5.05





6% H202 2.72 2.44 2.6 2.64

9% H2O2 Commercial 6.17 5.57 5.79 5.48





6% H202 3.02 2.89 2.86 2.64

9% H2O2 Commercial 5.32 5.61 4.96 5.23


Figures 6.14- 6.16 investigate the oxidation effects on bleached hair sample at the

shortest oxidation time (0.5h), for the middle stage of the oxidation times (2h) and for

the maximum oxidation time (4h). Since the valley depth of cysteic acid of commercial

persulphate bleached hair is always higher than that of the 6% H2O2 oxidation

treatment, indicates that more cystine is oxidised to cysteic acid. This further confirms

that commercially persulphate bleach has a stronger oxidation impact on the keratin

than 6% H2O bleach.


195

It was also found for the three types of bleached hairs that the differences of valley

depth for cysteic acid between the different parts of the hair fibres become smaller

with longer bleaching time. This shows that there is a uniform and homogenous

bleaching effect on the hair length from tip to root. Extensive oxidation treatment thus

reduces the variability of hair properties along length. However, small differences in

absorption intensities for each tress of hair still exist. This is likely because the

pressure applied to the hair fibres by the pressure tower of the ATR cell is different

every time.

Although different points on hair fibres have different amounts of oxidised cystine, the

oxidation effect yielded by different oxidizing agents is still our main interest.

Therefore, all of the valley depths results were combined to undertake an overall

oxidation damage comparison.


FTIR-ATR after bleaching time of 0.5h. The graph shows the mean and the standard

error for the cysteic acid absolute valley depth at 5 different measurements points for

three bleached hairs. The absolute valley depth is expressed in arbitrary units.

0

1

2

3

4

5

6

7

8


Cys

teic

aci

d a

bso

lute

val

ley

de

pth

aft

er

seco

nd

de

riva

tive

(ar

bit

rary

un

its)

Distance from root of hair samples

Bleaching time 0.5h

6% H202 Bleach




196


FTIR-ATR after bleaching time of 2h. The graph shows the mean and the standard




FTIR-ATR after bleaching time of 4h. The graph shows the mean and the standard



0

1

2

3

4

5

6

7

8


Cys

teic

aci

d a

bso

lute

val

ley

dep

th a

fte

r se

con

d d

eriv

ativ

e (

arb

itra

ry u

nit

s)

Distance from root of hair sample

Bleaching time 2h

6% H202 Bleach



0

1

2

3

4

5

6

7

8


Cys

teic

aci

d a

bso

lute

val

ley

dep

th a

fter

se

con

d d

eriv

ativ

e (a

rbit

rary

un

its)

Distance from root of hair samples

Bleaching time 4h 6% H202 Bleach




197

6.4.2.1.2 Effects of the Oxidizing Agents

As has been discussed in Section 6.3.1.2, it is sensible to apply spectral averaging prior

to peak height calculation, since hair is a highly heterogeneous material (Signori,

1997). A selected region of a spectrum is normalized against the intensity of CH2 at

1452cm-1

. The absorbance value of the absolute valley depth of cysteic acid is

calculated and compared to carry out quantitative analysis (using arbitrary units of

measurement) after second derivative analysis. After combining all of the results from

each measured point on a hair tress, Figure 6.17 is achieved according to the procedure

described above to demonstrate the effects of different oxidizing agents.

In Figure 6.16, commercial persulphate bleach and 9% H2O2 commercial bleach have a

stronger oxidation effect on cystine than 6% H2O2 bleach, which only has a moderate

oxidation influence. This observation is in a good agreement with the colour

measurements described in Chapter 4. Although all three oxidation treatments

demonstrate a consistent, increasing tendency towards equilibrium, 6% H2O2 bleach

increases steadily for the whole treatment time, whereas values for both 9% H2O2

commercial bleach and commercial persulphate bleach increase rapidly at the

beginning until they reach an equilibrium level. For 9% H2O2 commercial bleach the

equilibrium is reached at 2h and for commercial persulphate bleach is reached at 3.5h.

These two commercial bleach agents are formulated differently. 9% H2O2 commercial

bleach is in the form of a creamy solution, whereas commercial persulphate bleach is

prepared as a paste. Since the ATR technique is used to examine the surface of the hair

samples, the creamy solution facilitates the penetration of further layers of the cuticle

or of the cortex compared to the paste state of the commercial persulphate bleach. Thus,

9% H2O2 commercial bleach reaches its maximum value for cysteic acid intensity

earlier than the commercial persulphate bleach.


198

Figure 6.17: Oxidation effects on hairs measured using FTIR-ATR at various

oxidizing agents. Valley of cysteic acid after second derivative is shown at 1040 cm-1

.

The trendline curve has been calculated by a polynomial relationship.

Since two types of virgin hair sources were employed to perform the oxidation

treatments, such as hair batch I was used for 6% H2O2 bleach and 9% H2O2

commercial bleach, and hair batch II was adopted for commercial persulphate bleach,

their relative value of absolute valley depth of cysteic acid (Hgt R

= (Hgt/Hgt

0), where

Hgt0

represents the absolute valley depth of cysteic acid of virgin hair, were introduced

and were normalized against the internal standard. Figure 6.18 presents the comparison

of the relative values of valley depth of cysteic acid for the different oxidizing agents.

Here the relative values for 9% H2O2 commercial bleach are very dominant, which

were expected to take a middle place between commercial persulphate bleach, which is

the stronger oxidizing agent (persulphate), and 6% H2O2 bleach, which shows the mild

damage effect all the time. This observation is related to the “Knot Test” results

(Chapter 3) that 9% H2O2 commercial bleach induces the severer surface damage,

which leaves the cortex exposed at extreme bleaching conditions. However, 6% H202

bleach and commercial persulphate bleach lead to some remaining cuticle or has only

one layer of cuticle, which is less damaged than for the 9% H2O2 commercial bleach.

The values of the standard error SE, which arise from each treatment are too small to

be visible. This indicates that a relatively consistent bleaching effect produced on each

hair sample along the fibre length.

0

1

2

3

4

5

6

7

0 1 2 3 4

Cys

teic

aci

d a

bso

lute

val

ley

de

pth

aft

er

seco

nd

de

riva

tive

(ar

bit

rary

un

its)

Bleaching time (h)



6% H202 Bleach


199

Figure 6.18: Oxidation effects on bleached hair measured using FTIR-ATR at various

oxidizing agents. Relative values were employed, as Hgt R

= (Hgt/Hgt

0), where Hgt

0

represents the absolute valley depth of cysteic acid of virgin hair. The trendline curve

has been calculated by polynomial relationship.

6. 4. 2.2 KBr Transmission Mode

6.4.2.2.1 Effects of the Oxidizing Agents

The IR transmission technique, as a widely used routine method of FTIR spectroscopy,

was also utilised to detect the oxidation conditions of cystine in bleached hair keratin.

When the light is transmitted through the keratinous sample, spectral information is

provided not only from the surface layer, such as cuticle, but also from the inner part

of the fibre, the cortex. Although the reflectance mode of ATR can detect as deeply as

the evanescent wave penetrates in the cuticle (~2μm), it is not effective to analyse the

internal properties of hair. Consequently, transmission spectroscopy in the form of the

KBr pellet sampling technique was employed to investigate the internal chemistry

changes caused by oxidation in the cortex.

In Figure 6.19, the normalized FTIR spectra of untreated and treated hairs, which were

standardized against the peak heights at 1452 cm-1

clearly display the prominent peak

0

1

2

3

4

5

6

7

8

0 1 2 3 4

Rel

ativ

e cy

ste

ic a

cid

ab

solu

te v

alle

y d

epth

af

ter

seco

nd

der

ivat

ive

(ar

bit

rary

un

its)

Bleaching Time (h)

6% H202 Bleach




200

of cysteic acid at 1040 cm-1

, and secondly at 1172 cm-1

with the prolonged treatment

time (3h). The reason for choosing 3h treatment time as the terminal bleaching time is

because the 9% H202 commercial bleached hairs and commercial persulphate bleached

hairs were rinsed with water overnight (see Chapter 2). From the results of Chapter 5,

it is clear that the cross-linked matrix, which contributes most of the cystine in the

cortex, is highly affected by prolonged rinsing with water. We thus exclude the 3.5h

and 4h oxidation period for this internal property analysis.

The second cysteic acid peak at 1172 cm-1

is not obvious as shown in Figure 6.19 for

the 6% H202 treated and untreated hair samples. This indicates that these two types of

hair samples are subject to less oxidation damage than for 9% H202 commercial bleach

and commercial persulphate bleach. 9% H202 commercial bleach achieved the

strongest absorbance intensity for the cysteic acid peak among the three types of

bleached hair samples. This corresponds to the previous ATR results, even though the

commercial persulphate bleach contains the intensive oxidizer, persulphate. The reason

for this abnormal result is explained by the DSC result in Chapter 5, in that the matrix

for 9% H202 commercial bleach is heavily damaged after the intensive oxidation time

(≥2h), thus resulting in the abundance of cysteic acid.

Figure 6.19: FTIR-KBr spectrum of untreated and 3 types of 4h treated fibers at 1300

– 900cm-1

after normalization. Cysteic acid peak becomes clearly visible at 1040cm-1

and 1171 cm-1

for the bleached fiber, (marked by the red arrows). Each curve is

randomly chosen from one of three tresses’ results.

0.7

0.75

0.8

0.85

0.9

0.95

1

1.05

1.1

900 1000 1100 1200 1300

Ab

sorb

ance

aft

er

no

rmal

izat

ion

Wave number (cm-1)

6% H202 Bleach 3h

9% H202 Commercial Bleach 3h

Commercial Persulpahte Bleach 3h

Virgin Hair


201

After second derivative transformation, the obtained absorbance of valley depths for

cysteic acid at 1042 cm-1

were calculated and compared to illustrate the transformation

of the chemical structure of cystine to cysteic acid, caused by the three types of

bleaching treatments.

The second derivative spectra of untreated and three types of bleached hairs for 3h

bleaching time are shown in Figure 6.20. It is noticed that the cysteic acid valley

appears at 1040cm-1

. 9% H2O commercial bleach (3h) exhibits the most intensive

damage impact, followed by commercial persulphate bleach (3h), 6% H2O bleach (3h)

and virgin hair.

In agreement with the observations in Figure 6.13, the peak location after second

derivative moves slightly for the peak of cysteic acid from 1042 to 1037cm-1

. The

reason for this transformation is explained by the Hogg (1994) that partial disulphide

bond rupture causes the disulphide conformation.

Figure 6.20: FTIR-KBr spectrum of untreated and 3 types of 4h treated fibres at 1300

– 900cm-1

after second derivative. Cysteic acid becomes clearly visible as a valley at

1040cm-1

for the bleached fiber, (marked by the red arrow). Each curve is randomly

chosen from one of three tresses’ results.

It has been shown that the curves for 6% H2O2 and commercial persulphate bleach

behave in a similar way in Figure 6.21. They both increase during the first hour and

then remain constant. This reveals a homogenous oxidation effect on the matrix in the

-0.0014

-0.0012

-0.001

-0.0008

-0.0006

-0.0004

-0.0002

0

0.0002

0.0004

0.0006

0.0008

980 1030 1080 1130 C

yste

ic a

cid

val

ley

dep

th a

fter

sec

on

d

der

ivat

ive

(Arb

itra

ry u

nit

s)

Wavenumber (cm-1)

6% H202 Bleach 3h

9% H202 Commercial Bleach 3h

Commercial Persulphate Bleach 3h

Virgin Hair


202

cortex over the treatment time, which corresponds to the conclusions in Chapter 5.

Thus these results, which both investigate the internal and surface properties of the

human hair, will be compared in Chapter 7 for a better explanation.

The curve for 9% H2O2 commercial bleach shows an upward tendency towards its

saturation, however it seems that its equilium level has not been reached during three

hours of oxidation.

Figure 6.21: Oxidation effects on bleached hair measured using FTIR-KBr for various

oxidizing agents. The trendline curve has been calculated by polynomial relationship.

Though two different hair sources were used in the sample preparation, the relative

values of peak intensity for cysteic acid of virgin hair batch I and II are shown in

Figure 6.22 to yield comparable curves for different oxidizing agents after

normalization. It is obvious that 6% H2O2 bleach and commercial persulphate bleach

have a similar, almost overlapping steady trend line while the 9% H2O2 commercial

bleach is increasing during the time of the oxidation treatment. This implies that 9%

H2O2 commercial bleach has a stronger oxidation effect on the matrix during 3h

bleaching. The cysteic for both 6% H2O2 bleach and commercial persulphate bleach

are coinciding, which is unexpected because there is an aggressive oxidiser agent in

the commercial persulphate bleach. The “Knot Test” result could provide a good

explanation that both 6% H2O2 bleach and commercial persulphate bleach has a

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 0.5 1 1.5 2 2.5 3

Cys

teic

aci

d a

bso

lute

val

ley

de

pth

aft

er

seco

nd

de

riva

tive

( A

rbit

rary

un

its)

Bleaching Time (h)

6% H202 Bleach




203

moderate surface damage on the cuticle, since cuticle is used to protect the inner layers

of hair, there is a similar oxidation effect on the matrix on the cortex by 6% H2O2 and

commercial persulphate bleach.

Figure 6.22: Oxidation effects on bleached hair measured using FTIR-KBr at various

oxidizing agents. Relative values were employed, as Hgt R

= (Hgt/Hgt

0), where Hgt

0

represents the absolute valley depth of cysteic acid of virgin hair. The trendline curve

has been calculated by polynomial relationship.

6.5 Conclusion

Transmission FTIR spectroscopy detects the internal region of the hair, while the

FTIR-ATR technique analyses surface properties only. Thus, a comprehensive analysis

of cysteic acid was conducted by using these two methods.

9% H2O2 commercial bleach demonstrates high absorbance intensity for cysteic acid in

both FTIR-ATR and transmission, which was supposed to take the middle place

between commercial persulphate bleach and 6% H2O2 bleach. The reason for this

abnormal result is explained by the SEM and DSC results, which show that the hair

surface of 9% H2O2 commercial bleach undergoes severer damage, which leaves the

cortex exposed and in extreme poor condition. The intermediate filament for 9% H2O2

commercial bleach is heavily damaged after intensive oxidation treatment (≥2h).

Relatively, both 6% H2O2 bleach and commercial persulphate bleach are subject to

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

0 0.5 1 1.5 2 2.5 3

Re

lati

ve C

yste

ic a

cid

ab

solu

te v

alle

y d

ep

th

afte

r se

con

d d

eri

vati

ve (

Arb

itra

ry u

nit

s)

Bleaching Time (h)

6% H202 Bleach




204

very similar oxidation effects on the cortex. It is found in the “Knot Test” that 6%

H2O2 bleach and commercial persulphate bleach have a moderate surface damage for

the cuticle. Since the cuticle is used to protect the inner structures of hair, there is a

similar oxidation effect on the matrix in the cortex by 6% H2O2 and commercial

persulphate bleach.

As shown in Section 6.4.1, Attenuated Total Reflectance (ATR) spectra measured by

FTIR are similar to those collected by transmission measurements. Thus the

relationship of two FTIR-modes can be compared after normalisation. The results are

presented in Figure 6.23.

It is obvious that 6% H202 bleached hair is subject to a milder effect on both surface

and internal properties. This is reflected in the ATR and transmission results. A

consistent progressive damage effect is observed for 9% H202 commercial bleached

hair for both cuticle and cortex layers, indicating that abundant cysteic acid is formed

during the whole treatment time. Commercial persulphate bleach, containing the

stronger oxidising agent, has an unexpected result, in that it has a less surface damage

than 9% H2O2 commercial bleached hair in FTIR-ATR test, and a similar oxidation

effect on the matrix as 6% H202 bleached hair in FTIR transmission mode investigation.

Figure 6.23: Oxidation effects on bleached hair measured using FTIR-ATR and FTIR-

transmission mode for various oxidizing agents. Relative values were employed, as Hgt R

= (Hgt/Hgt

0), where Hgt

0 represents the absolute valley depth of cysteic acid of virgin

hair. The trendline curve has been calculated by polynomial relationship.

0.5

1.5

2.5

3.5

4.5

5.5

6.5

7.5

8.5

9.5

0 0.5 1 1.5 2 2.5 3

Re

lati

ve C

yste

ic a

cid

ab

solu

te v

alle

y d

ep

th

afte

r se

con

d d

eri

vate

(A

rbit

rary

un

its)

Bleaching Time (h)

ATR 6%H202 Bleach

ATR 9% Commercial Bleach

ATR Commercial Persulphate Bleach

KBr 6% H202 Bleach

KBr 9% Commercial Bleach

KBr Commercial Persulphate Bleach


205

6.6 References

ALTER, H., MICHAEL, BIT -ALKHAS., 1969. Infrared Analysis of Oxidized

Keratin. Textile Research Journal, , 39, 479-481.

BADDIEL, C. B. 1968. Structure and reactions of human hair keratin: an analysis by

infrared spectrsocopy. Molecular biology, 38, 181-199.

BARBARA, H. S. 2004. Infrared spectroscopy: fundamentals and applications,

England, John Wiley&Sons.

BARTH, A. 2007. Infrared spectroscopy of proteins. Biochimica et Biophysica Acta,

1767, 1073-1101.

BARTICK, E. G., TUNGOL, MARY W., REFFNER, JOHN A., 1994. A new

approach to forensic analysis with infrared microscopy: Internal reflection

spectroscopy. Analytical Chimica Acta, 288, 35-42.

BEER, M., SUTHERLAND, G. B. B. M., TANNER, K. N., AND WOOD, D. L., 1959.

Infrared Spectra and Structure of Proteins. Proceedings of the Royal Society,

147-172.

BELLAMY, L. J. 1958. The infrared Spectra of Complex Molecules, New York,

Wiley.

BELLAMY, L. J. 1975. The Infrared Spectra of Complex Molecules, New York,

Wiley.

BENÍTEZ, J., MATAS, AJ., HEREDIA A., 2004. Molecular characterization of the

plant biopolyester cutin by AFM and spectroscopic techniques. J Struct Biol,

147, 179-184.

BILINSKA, B., KOLCZYNSKA-SZAFRANIEC, U., BUSZMAN, E., 1991. The

structure of melanin biopolymers from human hair. I. IR spectroscopy studies.

Postepy Fizyki Medycznej, 26, 75-80.

BRENNER, L., SQUIRES, P. L., GARRY, M., AND TUMOSA, C. S., 1985. A

Measurement of Human Hair Oxidation by Fourier Transform Infrared

Spectroscopy. Journal of Forensic Sciences, 13, 420-426.


206

BRIDGE, T. P., FELL, A. F., WARDMAN, R. H., 1987. Perspective in Derivatives

Spectroscopy, Part 1: Theoretical Principles. Journal of the Society of Dyers

and Colourists, 103, 17-27.

CAROTI, G., DUSENBURY, JOSEPH H., 1956. Effect of drawing on the infrared

dichroism of nylon 66 filaments. Polymer Science, 399-407.

CARR, C. M., LEWIS, D. M. 1993. An FTIR spectroscopic study of the

photodegradation and thermal-degradation of wool. Journal of the Society of

Dyers and Colourists,, 109, 21-24.

CHURCH, J. S., CORINO, G. L., WOODHEAD, A. L., 1997. The analysis of Merino

wool cuticle and cortical cells by Fourier transform Raman spectroscopy.

Biopolymers, 42, 7-17.

COLTHUP, N. B., DALY, LAWRENCE H., WIBERLEY, STEPHEN E., 1964.

Introduction to Infrared and Raman Spectroscopy, New York, Academic Press.

CROOK, A., TAYLOR, P. J., 1958. Chemistry and industry, 95.

DEVANEY, R. G., THOMPSON, AVON L., 1958. An Improved Technique for

Preparasiong Samples of Organic Solids for Infrared Analysis. Applied

Spectroscopy, 12, 154-155.

FREDLINE, V., KOKOT, S., GILBERT, C., 1997. An FT-IR and Raman

spectroscopy study of electrochemically modified woollen fabric.

Mikrochimica Acta, 183-184.

GNIADECKA, M., NIELSEN, O. F., CHRISTENSEN, D. H., WULF, H. C., 1998.

Structure of water, proteins, and lipids in intact human skin, hair, and nail. J

Invest Dermatol, 110, 393-398.

GRDADOLNIK, J. 2002. ATR-FTIR Spectrscopy: its advantages and limitations. Acta

Chim. Slov, 49, 631-642.

HARRICK, N. J. 1979. Internal Reflection Spectroscopy, Ossining, New York,

Harrick Scientific Corp.

HARRICK, N. J. 1985. Internal Reflectance Spectroscopy, New York, Interscience

Publishers.

HARRIS, D. C. 1999. Quantitative Chemical Analysis USA, W. H. Freeman & Co Ltd.

HILTERHAUS-BONG, S., ZAHN, H., 1987. Contributions to the chemistry of human

hair. I. Analyses of cystine, cysteine and cystine oxides in untreated human hair.

International Journal Cosmetic Science, 9, 101-110.


207

HOGG, L. J., EDWARDS, H. G. M., FARWELL, D. W., AND PETERS, A.T., 1994.

FT-Raman Spectroscopic Studies of Wool. Society of Dyers Colourists, 110,

196-199.

HSU, C. P. S. 1997. Infrared Spectroscopy, New Jersey, Prentice Hall.

ITOU, T., NOJIRI, M., OOTSUKA, Y., NAKAMURA, K., 2006. Study of the

interaction between hair protein and organic acid that improves hair-set

durability by near-infrared spectroscopy. J Cosmet Sci, 57, 139-151.

JOY, M., LEWIS, D. M., 1991. The use of Fourier transform infrared spectroscopy in

the study of the surface chemistry of hair fibres. International Journal of

Cosmetic Science, 13, 249-261.

KHAN, J. I., KENNEDY, THOMAS J., CHRISTIAN, JR DONNELL R., 2012. Basic

Principles of Forensic Chemistry, Springer, Humana Press.

KIRSCHENBAUM, L. J., COUTINHO, L., HUFFMAN, S. W., BROWN, C. W.,

2000. Assessment of Strctural Changes in Human Hair by Near IR

Spectroscopy (Abstract). Abstract of Papers of the American Chemical Society.

KUZUHARA, A. 2005. Analysis of structral change in keratin fibers resulting from

chemical treatment using Raman spectrscopy. Biopolymers, 77, 335-344.

LEWIS, D. M., YAN, G., 1993. The effect of various chromium species on wool

keratin. Journal of the Society of Dyers and Colourists, 109, 193-198.

LYMAN, D. J., MURRAY-WIJELATH, JACQUELINE., FEUGHELMAN, MAX.,

2001. Effect of temperature on the conformation of extended alpha-keratin.

Applied Spectroscopy, 55, 552-554.

MARTIN, K. 1999. Cosmetic Spectroscopy: A review of infrared and Raman studies

of skin and hair. The Internet Journal of Vibrational Spectroscopy, 3, 1.

OHNISHI, K., SAKANO, T., FUJIWARA, N., 1993. Application of FTIR diamond

cell method to discrimination of personal hair. Journal of Toxicology and

Enviromental Health, 39, 247-250.

PAINER, P., STARSINIC, M. AND COLEMAN, M. 1985. Fourier Transform

Infrared Spectroscopy, Orlando, Academic Press.

PAINTER, P. C., COLEMAN, M.M., JENKINS, R.G., WHANG, P.W. AND

WALKER, P.L. 1978. Fourier transform infrared study of mineral matter in

coal: a novel method for quantitative mineralogical analysis. Fuel, 57, 337-344.


208

PAINTER, P. C., RIMMER, S. M., SNYDER, R. W., DAVIS, A 1981. A Fourier

transform infrared study of mineral matter in coal: the application of a least-

squares curve-fitting program. Applied Spectrsocopy, 35, 102-106.

PANAYIOTOU, H., KOKOT, S., 1999. Matching and discrimination of single human-

scalp hairs by FT-IR micro-spectroscopy and chemometrics. Analytica Chimica

Acta, 392, 223-235.

PAVIA, D. L., LAMPMAN, GARY M., KRIZ, GEORGE S., 2001. Introduction to

Spectroscopy, USA, Harcourt College Publishers.

PAVIA, D. L., LAMPMAN, GARY M., KRIZ, GEORGE S., VYVYAN, JAMES A.,

1996. Introduction to spectroscopy, USA, Saunders College Publishing.

PIELESZ, A., WKOCHOWICZ, A., BINIAS, W., 2000. The evaluation of structural

changes in wool fibre keratin treated with azo dyes by Fourier Transform

Infrared Spectroscopy. Spectrochimica Acta Part A-Molecular and

Biomolecular Spectroscopy, 56, 1409-1420.

ROBBINS, C. R. 1967. Infrared Analysis of Oxidised Keratins. Textile Research

Journal, 37, 811-813.

ROBBINS, C. R. 1994. Chemical and Physical Behavior of Human Hair, Berlin,

Springer.


Springer-Verlag.

ROBINSON, V. N. E. 1976. A Study of Damged Hair Journal of the Society of

Cosmetic Chemists, 27, 155-161.

SCHAARE, P. N., FRASER, D. G., 2000. Comparison of reflectance, interactance and

transmission modes of visible-near infrared spectroscopy for measuring

internal properties of kiwifruit (Actinidia chinensis). Postharvest Biology and

Technology, 20, 175-184.

SHAH, R. C., GANDHI, R. S., 1968. Infrared Analysis of Oxidized Keratins. Textile

Research Journal 38, 874.

SIGNORI, V., LEWIS, D. M., 1997. FTIR Invesigation of the damage produced on

human hair by weathering and bleaching processes: Implementation of

different sampling techniques and data processing. International Journal of

Cosmetic Science, 19, 1-14.

SILVERSTEIN, R. M., BASSLER, G. CLAYTON., MORRILL, TERENCE C., 1981.

Spectrometric Identification of Organic Compounds, New York, Wiley.


209

SIMON, A., KUNATH, D., 1961. Vibrational spectra and structure of s-alkyl

thiosulfates (Bunte salts). Ber. Deutche Chem. Ges, 94, 1980-1986.

SMITH, B. C. 1996. Fundamentals of Fourier transform infrared spectroscopy, USA,

CRC Press

STRASHEIM, A., BUIJS, K., 1961. An infrared study of the oxidation of the disulfide

bond in wool. Biochimica Biophysical Acta, 47, 538-541.

THERMONICOLET 2001. Introduction to Fourier Transform Infrared Spectrometry.

TOLGYESI, E. 1983. Weathering of hair Cosmetics and Toiletries, 98, 29-33.

VENYAMINOV, S. Y., PRENDERGAST, F. G., 1997. Water (H20 and D20) molar

absorptivity in the 1000-4000 cm-1 range and quantitative infrared

spectroscopy of aqueous solutions. Analytical Biochemistry, 248, 234-245.

WESTON, G. J. 1955. The infrared spectrum of peracetic acid-treated wool. Biochim.

Biophys. Acta, 47, 462-464.

WOLFRAM, L. J., LINDEMANN, M. K. O., 1971. Some observations on the hair

cuticle. Journal of the Society of Cosmetic Chemists, 22, 839-850.

Chapter 7 Summary and Conclusion

210


Evaluation of hair damage is particularly important in assessing the performance of

hair bleaches, which involves an oxidation reaction during the application. A variety of

methods was combined to correlate the surface and structural changes in human hair

imparted by the oxidation treatments. They are: Scanning Electron Microscopy (SEM),

Reflective Spectrophotometry, Differential Scanning Calorimetry (DSC) and Fourier

transform infrared spectroscopy (FTIR). Different levels of damage were achieved by

changing the time of treatment up to 4 hours for 6% H202 (hydrogen peroxide) as well

as commercial bleaches, such as 9% H202 commercial bleach and commercial

persulphate bleach (contains 9% H202).

The morphological changes of the cuticle as a consequence of damage processes are

investigated by means of SEM. When hair fibres are subject to bleach damage, the

cuticle is progressively destroyed, resulting in the cuticle edges being gradually

chipped away. Under severe treatment conditions, the layer of cuticular sheets is

entirely removed, leaving the underlying cortex exposed. A systematic numerical

approach named “Knot Test” as a new methodology is applied using cuticle

descriptors to evaluate hair damage. There are two types of damage classifications:

mild and heavy damage classifications. The “Knot Test” results show that surface

changes increase with peroxide treatment time. The longer treatment time results in the

greatest surface damage. Surprisingly, commercial persulphate bleaching treatment

produced the least topographical damage with the extended treatments among all the

bleaching conditions. 6% H202 bleach has an overall moderate damage effect over the

treatment time. 9% H202 commercial bleach shows a distinctive damage result. For the

first 1.5h bleached hairs show mild oxidative damage, while severe damage occurs

after 2h.

Colour changes of bleached hair with increasing treatment time are evaluated, since the

initial response of oxidation to the hair fibre is colour change. In general, bleached

hairs have an overall lighter, yellowish and reddish colour change. The lightness

change of the treated hair is due to pigment oxidation. Yellowness change is associated


211

to two steps, namely, solubilisation of black-brown melanin and keratin degradation. It

has been verified that a* (redness) of hair plays a minor role in determining the colour

change, so that b* (yellowness) and L* (lightness) are the main controlling

components in the CIE L*a*b* colour space, which is an objective tool for the

evaluation of colour specification. The higher the concentration of peroxide and the

longer the treatment time, the larger is the ∆E* (total colour change) value, so the

greater is the hair colour difference. 9% H202 commercial bleach and especially the

commercial persulphate bleach demonstrate an overall larger increase in colour change

than the 6% H202 bleach, which always has a mild oxidation effect. Consequently,

commercial persulphate bleached hair is much lighter and more yellow than 9% H202

commercial bleached hair and 6% H202 bleached hair.

Thermal analysis (DSC) investigates the structural changes and degree of degradation

of the cortex based on the denaturation enthalpy ∆HD and denaturation temperature TD

of keratin. These parameters reflect the amount and structural integrity of the α-helical

material in the IFs, and the cross-link density of the matrix (IFAPs), respectively. There

is a consistent decrease in both TD and ∆HD by the three bleaching treatments over the

bleaching time, showing that the structural damage becomes greater with the longer

treatment time. The deconvolution results for the three bleaches using three Gaussian

distributions confirm this observation. The stronger bleach results in a homogenous

structural damage on both para- and ortho-cortex with prolonged bleaching time.

Commercial persulphate bleach and 9% H202 commercial bleach have a larger

progressive damage effect on the ortho- and para- cortex than 6% H202 bleach.

A strong linear relationship between ∆HD and TD was observed for 6% H2O2 bleach,

commercial persulphate bleach and for parts of the data for 9% H202 commercial

bleach, revealing that the three bleaches have a homogenous bleaching effect on IFs

and IFAPs. However, deviating results for 9% H202 commercial bleach after 2 hours

bleaching time are observed. This is because the intermediate filament is progressively

damaged after excessive bleaching treatment (≥2h), so that there is a larger decrease in

∆HD than in TD. “Knot test” results show the cuticle surfaces of 9% H202 commercial

bleach to be strongly degradated after 2h. As a consequence, 9% H202 commercial

bleach (2h) has a severe oxidative damage effect on both cuticle and cortex of hairs.


212

Kinetic analysis is conducted for virgin and commercial persulphate bleached hairs by

combining data for the various heating rates according to ASTM-E698. It is confirmed

that the denaturation temperature shifts to higher values with increasing heating rate.

The activation energies of 260 kJ/mol for the virgin hair and 295 kJ/mol for the

commercial persulphate bleached hair (2h) are determined from the slope of the linear

regression between the non-reversing denaturation peak temperature (as 1/TD) and

heating rate, β (as lnβ) on the basis of the Arrhenius-equation. Since the difference of

their activation energies is quite small (35 kJ/mol), there is no obvious damage change

in the matrix, thus the predominant structural damage by heating rate only occurs in

the IF. The relationship between denaturation enthalpy ∆HD and heating rate provides a

better explanation for this finding. A linear increase in ∆HD for lower heating rates is

observed between 0.15 ~ 1oC/min, while it is constant for higher heating rates (up to

7oC/min). This can be ascribed to the hypothesis that a lower heating rate favours a

crystal transformation change (α-β transformation), while a higher rate favours a

crystalline-amorphous transformation. Virgin mohair is also compared with virgin hair

using this approach. The slope of linear regression for the lower heating rates increase

from mohair to human hair, however they overlap at the higher heating rates,

indicating that both virgin keratin fibres have similar amounts of “native” α-helix.

SEM has been used to examine the morphological changes of the denaturated hair

samples after DSC. There is a common characteristic that the cortex is dissolved at

lower heating rates. The commercial persulphate bleached hairs show an overall

shrunk cuticle surface and fewer and smaller hydrolysed protein granules, due to the

previously damaged α-helix in the cortical cell.

Although commercial persulphate bleached hair (2h) has an overall lower denaturation

temperature for each cortical cell type for all heating rates, the distance between the

para- and ortho- cortex is the same as for virgin hair. This shows that the heating rate

affects the cystine linkages of para- and ortho- cortex homogenously, regardless of the

previous damage. However, a homogenous damage on the three cortical components

can only occur for the previously damaged fibres.

Reflectance and transmission are two modes of FTIR spectroscopic measurements.

Since the cuticle surfaces and the matrix of 9% H202 commercial bleach are subjected


213

to the severe damage after prolonged oxidation treatment (2h), as confirmed by SEM

and DSC, 9% H202 commercial bleach shows the highest absorbance intensity for

cysteic acid peak in both FTIR-ATR and transmission. Both 6% H2O2 and commercial

persulphate bleach show a similar oxidation effect in the FTIR transmission

investigations. This findings are interpreted such that the 6% H2O2 and commercial

persulphate bleach have a moderate surface damage on the cuticle, which protects the

cortex of hair, thus a similar oxidation effect on the cortex is achieved. Although

commercial persulphate bleach contains the stronger oxidising agent, it has a less

severe surface damage than 9% H202 commercial bleached hair as shown by the FTIR-

ATR measurements, and the similar oxidation effect on the matrix for 6% H202

bleached hair as shown by the FTIR transmission investigation.

7.1 Comparison of FTIR-KBr and DSC Results

Since both FTIR-KBr and DSC are detecting internal properties of the cortex, the

comparison of relative cysteic acid absolute valley depth Hgt R, Hgt

R =

(Hgt/Hgt

0) in the

FTIR transmission mode, against the values of relative denaturation temperature TD,

TD R

= (TD/TD

0) in DSC measurement is given in Figure 7.1. One phenomenon, which

draws attention is that the data generated by 6% H202 bleach and 9% H202 commercial

bleach follow a regular decreasing pattern, while the results for commercial

persulphate bleach behave differently, which is unexpected.


214

Figure 7.1: Comparison of the relative cysteic acid absolute valley depth HgtR, Hgt

R =

(Hgt/Hgt0) from FTIR-KBr, with relative denaturation temperature TD, TD

R =

(TD/TD

0)

from DSC measurements.

Overall, 6% H2O2 has a moderate damage effect on both surface and structural change.

Commercial persulphate bleach causes much less surface damage than 9% commercial

peroxide bleach. Although structural changes increase with peroxide treatment time,

FTIR transmission mode found that 9% commercial peroxide treatment has stronger

damage effects on the cortex than the commercial persulphate bleach. This is

corroborated by DSC investigations, which show that the IF for 9% commercial bleach

is heavily damaged after the intensive oxidation treatment time (≥2h).

7.2 Future Work

This information provides more insight into how hair changes on the molecular level,

and in turn may lead to better treatment formulations or procedures with minimal

damage to the hair. In order to assess hair damage more systemically, it has been

suggested to combine the approach with other analysis, such as amino acid analysis,

dynamic mechanical analysis (DMA), gloss test, tensile test, nano-indentation and

others, which seem to be effective in identifying hair damage.

0.84

0.86

0.88

0.9

0.92

0.94

0.96

0.98

1

1.02

0 1 2 3 4 5

Re

lati

ve C

yste

ic a

cid

ab

solu

te v

alle

y d

ep

th

afte

r se

con

d d

eri

vati

ve (

Arb

itra

ry U

nit

s)

Relative Denaturation Temperature (Arbitrary Units)

6% H202 Bleach



Chapter 8 Appendix

215

Chapter 8 Appendix

Table 8.1: The reflectance data for 6% H202 bleached hair samples in the visible

spectrum

Wavelength

(nm)

0h 0.5h 1h 1.5h 2h 2.5h 3h 3.5h 4h

400 3.28 3.34 4.2 4.4 4.47 4.85 4.64 5.03 5.59

410 3.07 3.39 3.9 4.1 4.28 4.66 4.68 5.29 5.62

420 3.14 3.52 4 4.28 4.48 4.92 5 5.72 6.1

430 3.21 3.68 4.25 4.61 4.83 5.36 5.45 6.23 6.66

440 3.31 3.88 4.5 4.96 5.23 5.82 5.94 6.82 7.31

450 3.4 4.08 4.79 5.34 5.67 6.31 6.48 7.43 7.99

460 3.51 4.3 5.11 5.78 6.18 6.85 7.05 8.1 8.73

470 3.61 4.54 5.46 6.23 6.71 7.43 7.67 8.78 9.49

480 3.75 4.85 5.9 6.79 7.35 8.12 8.4 9.59 10.38

490 3.87 5.16 6.33 7.35 8.01 8.82 9.17 10.42 11.29

500 4.03 5.52 6.83 7.98 8.74 9.59 9.98 11.31 12.28

510 4.21 5.9 7.37 8.65 9.52 10.41 10.86 12.24 13.32

520 4.39 6.31 7.94 9.35 10.33 11.27 11.75 13.19 14.39

530 4.58 6.76 8.55 10.11 11.19 12.18 12.68 14.18 15.49

540 4.79 7.21 9.19 10.88 12.08 13.11 13.65 15.19 16.62

550 5.03 7.72 9.88 11.71 13.01 14.09 14.67 16.25 17.81

560 5.28 8.24 10.6 12.56 13.98 15.09 15.71 17.33 19

570 5.55 8.79 11.36 13.46 15 16.14 16.79 18.43 20.23

580 5.83 9.37 12.16 14.38 16.04 17.22 17.89 19.55 21.46

590 6.14 9.98 13 15.35 17.12 18.32 19.04 20.7 22.71

600 6.48 10.62 13.89 16.36 18.24 19.46 20.21 21.89 24.01

610 6.83 11.3 14.8 17.39 19.4 20.62 21.42 23.11 25.3

620 7.2 11.97 15.74 18.44 20.55 21.78 22.62 24.3 26.57

630 7.59 12.69 16.72 19.52 21.76 22.99 23.87 25.54 27.87

640 8.02 13.46 17.75 20.64 22.99 24.21 25.15 26.82 29.19

650 8.47 14.23 18.8 21.77 24.24 25.45 26.44 28.1 30.51

660 8.94 15.05 19.9 22.96 25.54 26.73 27.78 29.43 31.83

670 9.42 15.86 20.97 24.1 26.79 27.96 29.07 30.71 33.09

680 9.95 16.72 22.11 25.28 28.09 29.24 30.42 32.05 34.38

690 10.46 17.53 23.22 26.43 29.33 30.45 31.71 33.27 35.55

700 11.06 18.47 24.45 27.71 30.72 31.81 33.14 34.7 36.88

Chapter 8 Appendix

216

Table 8.2: The reflectance data for 9% H202 commercial bleached hair samples in the

visible spectrum

Wavelength

(nm)

0h 0.5h 1h 1.5h 2h 2.5h 3h 3.5h 4h

400 3.28 3.93 4.71 5.62 7.39 8.71 8.25 11.59 12.26

410 3.07 3.94 4.86 5.86 7.5 9.06 8.54 12.31 13.09

420 3.14 4.14 5.18 6.34 8.2 9.92 9.37 13.62 14.44

430 3.21 4.4 5.6 6.92 9.01 10.89 10.29 14.97 15.82

440 3.31 4.69 6.1 7.57 9.88 11.86 11.25 16.21 17.08

450 3.4 5.05 6.64 8.27 10.76 12.85 12.22 17.41 18.29

460 3.51 5.45 7.24 9.04 11.69 13.88 13.24 18.63 19.51

470 3.61 5.88 7.88 9.85 12.65 14.91 14.28 19.82 20.69

480 3.75 6.36 8.6 10.75 13.72 16.07 15.42 21.12 21.98

490 3.87 6.87 9.35 11.68 14.79 17.2 16.55 22.37 23.2

500 4.03 7.47 10.18 12.69 15.91 18.38 17.72 23.63 24.45

510 4.21 8.08 11.04 13.72 17.05 19.57 18.9 24.88 25.7

520 4.39 8.73 11.92 14.78 18.19 20.76 20.06 26.11 26.9

530 4.58 9.42 12.85 15.88 19.36 21.96 21.25 27.33 28.1

540 4.79 10.16 13.82 17.01 20.52 23.16 22.43 28.53 29.26

550 5.03 10.94 14.83 18.18 21.73 24.39 23.64 29.74 30.45

560 5.28 11.74 15.86 19.36 22.93 25.61 24.84 30.94 31.6

570 5.55 12.59 16.91 20.57 24.16 26.85 26.05 32.14 32.77

580 5.83 13.47 17.99 21.8 25.4 28.1 27.28 33.35 33.93

590 6.14 14.37 19.09 23.05 26.66 29.35 28.52 34.57 35.09

600 6.48 15.32 20.23 24.33 27.94 30.64 29.78 35.79 36.27

610 6.83 16.32 21.41 25.64 29.22 31.9 31.03 37 37.43

620 7.2 17.32 22.58 26.94 30.51 33.16 32.27 38.19 38.57

630 7.59 18.36 23.78 28.26 31.81 34.44 33.53 39.41 39.74

640 8.02 19.44 25.01 29.61 33.16 35.73 34.83 40.64 40.92

650 8.47 20.54 26.24 30.96 34.53 37.03 36.14 41.89 42.12

660 8.94 21.69 27.52 32.35 35.93 38.37 37.47 43.2 43.36

670 9.42 22.82 28.76 33.69 37.26 39.61 38.73 44.38 44.51

680 9.95 23.96 30.01 35.03 38.64 40.9 40.03 45.6 45.7

690 10.46 25.1 31.23 36.33 39.94 42.09 41.25 46.73 46.81

700 11.06 26.36 32.57 37.77 41.36 43.45 42.62 48 48.07

Chapter 8 Appendix

217

Table 8.3: The reflectance data for commercial persulphate bleached hair samples in

the visible spectrum

Wavelength

(nm)

0h 0.5h 1h 1.5h 2h 2.5h 3h 3.5h 4h

400 3.44 5.48 6.29 7.58 9.15 10.54 12.85 14.52 16.37

410 3.34 5.52 6.55 8 9.94 11.48 14.25 16.37 18.52

420 3.41 5.98 7.22 8.94 11.32 13.12 16.33 18.94 21.4

430 3.49 6.53 8.04 10.05 12.86 14.93 18.46 21.57 24.2

440 3.58 7.16 8.92 11.24 14.45 16.76 20.55 24.07 26.78

450 3.69 7.83 9.88 12.48 16.09 18.59 22.59 26.47 29.17

460 3.79 8.57 10.88 13.78 17.78 20.47 24.62 28.8 31.48

470 3.93 9.36 11.95 15.12 19.45 22.31 26.57 31 33.64

480 4.1 10.25 13.13 16.57 21.25 24.28 28.63 33.27 35.86

490 4.26 11.17 14.35 18.05 23.04 26.21 30.62 35.41 37.97

500 4.47 12.17 15.66 19.59 24.87 28.17 32.6 37.49 40.03

510 4.67 13.22 17 21.16 26.69 30.1 34.55 39.46 41.99

520 4.88 14.31 18.34 22.73 28.47 31.98 36.4 41.28 43.81

530 5.12 15.46 19.75 24.34 30.28 33.86 38.25 43.03 45.58

540 5.38 16.64 21.19 25.94 32.06 35.7 40.04 44.67 47.24

550 5.67 17.89 22.68 27.6 33.86 37.55 41.83 46.29 48.88

560 5.98 19.16 24.19 29.27 35.65 39.36 43.56 47.82 50.42

570 6.3 20.48 25.72 30.95 37.43 41.16 45.26 49.28 51.89

580 6.66 21.83 27.29 32.64 39.2 42.92 46.91 50.67 53.28

590 7.05 23.22 28.88 34.35 40.97 44.67 48.54 52.01 54.62

600 7.47 24.66 30.51 36.09 42.74 46.41 50.16 53.33 55.93

610 7.9 26.11 32.13 37.81 44.48 48.1 51.7 54.58 57.16

620 8.36 27.54 33.74 39.47 46.11 49.69 53.14 55.75 58.28

630 8.86 29.01 35.37 41.16 47.75 51.28 54.57 56.93 59.42

640 9.39 30.51 37.02 42.86 49.38 52.87 55.98 58.16 60.56

650 9.94 32.03 38.66 44.55 50.98 54.42 57.36 59.43 61.7

660 10.55 33.58 40.33 46.26 52.56 55.99 58.74 60.75 62.87

670 11.14 35.05 41.91 47.84 53.99 57.38 59.93 61.87 63.85

680 11.8 36.57 43.51 49.45 55.44 58.77 61.16 62.96 64.82

690 12.46 37.99 45.06 50.92 56.74 60.02 62.27 63.91 65.7

700 13.2 39.59 46.72 52.57 58.22 61.42 63.54 64.97 66.69

Chapter 8 Appendix

218

Table 8.4: Denaturation Temperatures and non-reversing Cp of virgin hair at various

heating rates for a typical DSC-curve

β

TD CpNR

0.5 1.5 2 2.5 3 4 5 7

80 0 0 0 0 0 0 0 0

81 13.12 0 4.17 2.685 4.07 0 0 0

82 13.52 0.9045 3.706 1.664 4.065 12.34 8.771 0

83 13.47 0.5624 3.289 1.792 3.909 13.96 9.602 5.011

84 13.5 0.3366 3.535 1.967 3.748 15.11 10.47 5.224

85 13.46 -0.3933 3.662 1.832 3.485 16.13 11.32 5.286

86 13.43 -0.1043 3.575 1.738 3.768 16.94 11.91 5.332

87 13.37 0.3505 3.526 1.701 3.824 17.46 12.35 5.387

88 13.36 0.698 3.487 1.662 3.77 17.42 12.51 5.407

89 13.35 0.6786 3.438 1.622 3.701 16.96 12.4 5.378

90 13.28 0.5829 3.39 1.586 3.678 16.63 12.24 5.326

91 13.28 0.5733 3.377 1.563 3.678 16.55 12.15 5.299

92 13.32 0.5642 3.362 1.55 3.664 16.63 12.13 5.322

93 13.31 0.5655 3.338 1.537 3.646 16.72 12.18 5.351

94 13.33 0.5634 3.315 1.525 3.625 16.79 12.22 5.336

95 13.31 0.5714 3.297 1.518 3.605 16.83 12.26 5.292

96 13.27 0.5736 3.282 1.518 3.584 16.86 12.29 5.249

97 13.27 0.5802 3.268 1.511 3.575 16.89 12.29 5.216

98 13.25 0.5787 3.253 1.496 3.568 16.92 12.29 5.175

99 13.28 0.5847 3.234 1.485 3.563 16.95 12.29 5.134

100 13.32 0.5928 3.211 1.478 3.559 16.97 12.29 5.115

101 13.3 0.5986 3.189 1.475 3.554 16.98 12.3 5.107

102 13.29 0.5955 3.167 1.468 3.549 16.99 12.31 5.095

103 13.25 0.59 3.142 1.459 3.542 16.99 12.31 5.063

104 13.26 0.5851 3.117 1.452 3.537 16.99 12.31 5.012

105 13.26 0.5779 3.092 1.45 3.533 16.98 12.31 4.968

106 13.27 0.5754 3.075 1.447 3.528 16.97 12.3 4.95

107 13.27 0.571 3.066 1.442 3.526 16.97 12.3 4.936

108 13.28 0.5737 3.061 1.435 3.521 16.96 12.29 4.914

109 13.26 0.5761 3.047 1.433 3.512 16.95 12.29 4.866

110 13.27 0.5744 3.012 1.429 3.508 16.95 12.29 4.796

111 13.29 0.5824 2.983 1.428 3.505 16.95 12.28 4.724

Chapter 8 Appendix

219

112 13.3 0.582 2.975 1.425 3.504 16.94 12.27 4.653

113 13.35 0.5879 2.96 1.428 3.498 16.94 12.27 4.567

114 13.44 0.5893 2.937 1.432 3.492 16.94 12.26 4.475

115 13.38 0.5882 2.916 1.429 3.491 16.93 12.26 4.393

116 13.41 0.5799 2.896 1.432 3.493 16.93 12.25 4.337

117 13.47 0.5765 2.867 1.439 3.497 16.94 12.25 4.32

118 13.48 0.577 2.833 1.449 3.498 16.94 12.24 4.309

119 13.51 0.5759 2.815 1.457 3.496 16.95 12.24 4.281

120 13.5 0.5699 2.8 1.456 3.497 16.95 12.24 4.254

121 13.55 0.5634 2.775 1.457 3.498 16.94 12.24 4.23

122 13.58 0.5667 2.747 1.462 3.504 16.94 12.24 4.212

123 13.6 0.5585 2.718 1.462 3.505 16.94 12.24 4.214

124 13.67 0.559 2.683 1.452 3.505 16.94 12.23 4.228

125 13.75 0.5548 2.656 1.449 3.504 16.94 12.23 4.237

126 13.86 0.5635 2.639 1.454 3.504 16.93 12.22 4.236

127 13.93 0.5774 2.623 1.457 3.502 16.93 12.22 4.235

128 14.05 0.5955 2.608 1.462 3.501 16.93 12.21 4.237

129 14.22 0.6205 2.589 1.472 3.503 16.93 12.21 4.242

130 14.49 0.6567 2.582 1.494 3.513 16.92 12.2 4.249

131 14.77 0.7017 2.584 1.527 3.527 16.91 12.2 4.257

132 15.25 0.7837 2.601 1.566 3.552 16.91 12.19 4.265

133 15.69 0.8881 2.631 1.617 3.583 16.91 12.18 4.276

134 15.73 1.042 2.677 1.686 3.623 16.93 12.18 4.292

135 15.27 1.269 2.753 1.771 3.671 16.94 12.19 4.311

136 14.79 1.581 2.866 1.899 3.734 16.96 12.2 4.332

137 14.24 1.895 3.048 2.081 3.823 16.99 12.22 4.354

138 13.92 2.078 3.33 2.353 3.951 17.04 12.24 4.377

139 13.88 2.02 3.695 2.734 4.125 17.11 12.27 4.412

140 13.96 1.786 3.938 3.113 4.351 17.2 12.31 4.461

141 14.03 1.423 3.847 3.266 4.598 17.34 12.37 4.53

142 14.08 0.9998 3.521 3.087 4.786 17.53 12.45 4.618

143 14.28 0.5622 3.142 2.738 4.839 17.77 12.55 4.734

144 14.52 0.2006 2.701 2.387 4.74 17.98 12.68 4.876

145 14.7 -0.0167 2.212 2.051 4.555 18.12 12.83 5.031

146 15.03 -0.1299 1.828 1.714 4.311 18.12 12.99 5.171

147 15.45 -0.1754 1.651 1.412 4.024 18 13.11 5.275

Chapter 8 Appendix

220

148 15.68 -0.2051 1.587 1.218 3.72 17.83 13.17 5.329

149 15.69 -0.2483 1.548 1.154 3.477 17.63 13.15 5.319

150 15.82 -0.299 1.515 1.148 3.351 17.45 13.04 5.236

151 15.89 -0.3467 1.472 1.154 3.318 17.31 12.86 5.093

152 16.02 -0.3888 1.437 1.153 3.316 17.19 12.65 4.911

153 16.16 -0.4267 1.405 1.142 3.318 17.08 12.42 4.708

154 16.28 -0.4682 1.356 1.122 3.316 16.98 12.21 4.506

155 16.39 -0.4975 1.303 1.106 3.316 16.88 12.04 4.323

156 16.48 -0.5093 1.268 1.097 3.32 16.81 11.91 4.174

157 16.64 -0.4856 1.245 1.086 3.317 16.76 11.82 4.069

158 16.78 -0.4182 1.202 1.075 3.312 16.73 11.75 4.013

159 16.94 -0.2669 1.185 1.053 3.305 16.71 11.71 3.987

160 17.08 -0.0043 1.186 1.031 3.295 16.69 11.68 3.975

161 17.2 0.3595 1.214 1.015 3.282 16.68 11.66 3.969

162 17.32 0.8183 1.29 1.009 3.267 16.68 11.64 3.97

163 17.47 1.308 1.435 1.027 3.25 16.68 11.63 3.973

164 17.59 1.754 1.652 1.09 3.229 16.69 11.62 3.977

165 17.75 2.086 1.98 1.208 3.198 16.71 11.62

166 17.93 2.293 2.422 1.393 3.167 16.74 11.62

167 17.87 2.405 2.954 1.644 3.158 16.71 11.62

168 18.09 2.48 3.472 1.968 3.112 16.67 11.63

169 18.15 2.558 3.876 2.386 3.084 16.66 11.65

170 18.29 2.619 4.129 2.831 3.089 16.69

171 18.57 2.705 4.291 3.242 3.17 16.76

172 18.62 2.799 4.413 3.546 3.296

173 18.83 2.902 4.531 3.754 3.463

174 18.89 2.983 4.641 3.904

175 19.04 4.769

176 19.2

177 19.39

178 19.72

179

180

Chapter 8 Appendix

221

Table 8.5: Denaturation Temperature and non-reversing Cp of Commercial Persulphate

Bleached hair (2h) at various heating rates for a typical DSC-curve

β

TD CpNR

0.5 1 1.5 2 3 4 5 7

80 0 0 0 0 0 0 0 0

81 11.28 11.16 10.17 10.18 8.342 0 0 0

82 10.88 10.33 10.41 11.03 9.784 10.89 9.818 0

83 10.89 10.58 10.8 11.73 11.32 11.57 9.643 8.457

84 10.91 10.62 10.25 10.9 12.15 11.97 9.94 8.88

85 10.93 10.6 10.39 10.91 11.86 12.22 9.807 9.104

86 10.93 10.66 10.42 11.04 11.45 12.31 10.28 9.319

87 10.89 10.66 10.45 11.05 11.43 12.71 10.62 9.523

88 10.94 10.68 10.43 11.07 11.55 12.71 10.41 9.655

89 10.94 10.67 10.44 11.09 11.64 12.37 10.31 9.721

90 10.91 10.65 10.44 11.07 11.69 12.15 10.23 9.795

91 10.94 10.65 10.43 11.05 11.72 11.99 10.1 9.908

92 10.89 10.65 10.43 11.05 11.73 12.06 10.1 10.01

93 10.92 10.64 10.43 11.05 11.75 12.12 10.11 10.02

94 10.92 10.63 10.43 11.04 11.74 12.16 10.11 9.969

95 10.93 10.64 10.44 11.03 11.73 12.19 10.12 9.913

96 10.93 10.63 10.45 11.03 11.72 12.2 10.14 9.874

97 10.94 10.63 10.44 11.03 11.72 12.2 10.13 9.841

98 10.98 10.63 10.43 11.01 11.72 12.21 10.13 9.816

99 10.99 10.64 10.44 11 11.71 12.22 10.14 9.818

100 11 10.63 10.43 11 11.7 12.22 10.15 9.835

101 11 10.63 10.42 11 11.68 12.2 10.16 9.843

102 10.98 10.62 10.42 10.99 11.67 12.2 10.17 9.849

103 10.99 10.63 10.45 10.98 11.66 12.19 10.18 9.854

104 11.03 10.64 10.47 10.98 11.65 12.19 10.19 9.86

105 11.12 10.63 10.45 10.98 11.64 12.17 10.21 9.869

106 11.06 10.64 10.44 10.98 11.63 12.16 10.22 9.876

107 11.09 10.65 10.43 10.98 11.62 12.15 10.23 9.884

Chapter 8 Appendix

222

108 11.12 10.68 10.42 10.98 11.61 12.14 10.25 9.895

109 11.18 10.69 10.42 10.98 11.61 12.13 10.26 9.903

110 11.19 10.72 10.42 10.98 11.6 12.12 10.28 9.912

111 11.26 10.76 10.42 10.97 11.61 12.11 10.3 9.919

112 11.33 10.77 10.44 10.99 11.61 12.11 10.31 9.931

113 11.41 10.82 10.45 11 11.61 12.1 10.33 9.946

114 11.51 10.87 10.49 11.01 11.61 12.1 10.35 9.959

115 11.57 10.92 10.52 11.04 11.61 12.11 10.38 9.971

116 11.66 10.98 10.57 11.07 11.61 12.13 10.42 9.984

117 11.73 11.07 10.64 11.12 11.62 12.15 10.46 9.999

118 11.73 11.18 10.72 11.16 11.64 12.17 10.5 10.02

119 11.66 11.27 10.81 11.23 11.67 12.2 10.55 10.05

120 11.55 11.32 10.9 11.31 11.7 12.25 10.6 10.08

121 11.33 11.32 10.98 11.39 11.75 12.32 10.66 10.12

122 11.24 11.25 11.03 11.48 11.8 12.41 10.72 10.15

123 11 11.08 11.02 11.53 11.85 12.51 10.79 10.2

124 10.97 10.83 10.88 11.54 11.9 12.59 10.87 10.25

125 10.94 10.58 10.66 11.49 11.94 12.66 10.96 10.3

126 10.94 10.41 10.42 11.37 11.96 12.68 11.03 10.34

127 10.98 10.32 10.22 11.18 11.93 12.66 11.08 10.38

128 10.96 10.27 10.08 10.96 11.85 12.59 11.09 10.41

129 10.99 10.27 10.03 10.78 11.71 12.46 11.05 10.43

130 11.08 10.26 10.02 10.68 11.53 12.28 10.95 10.43

131 11.03 10.27 10.01 10.64 11.32 12.09 10.83 10.41

132 11.15 10.33 10.01 10.62 11.13 11.9 10.71 10.36

133 11.22 10.33 10.02 10.61 10.99 11.76 10.59 10.29

134 11.28 10.34 10.02 10.6 10.9 11.66 10.5 10.2

135 11.31 10.36 10.01 10.6 10.85 11.6 10.44 10.11

136 11.37 10.38 10.02 10.6 10.82 11.55 10.41 10.01

137 11.38 10.39 10.03 10.6 10.8 11.52 10.39 9.918

138 11.5 10.41 10.04 10.61 10.78 11.49 10.38 9.838

Chapter 8 Appendix

223

139 11.62 10.4 10.04 10.62 10.76 11.47 10.4 9.769

140 11.65 10.42 10.06 10.64 10.75 11.46 10.42 9.713

141 11.81 10.45 10.08 10.65 10.73 11.44 10.43 9.671

142 12.11 10.45 10.08 10.66 10.72 11.43 10.46 9.644

143 12.56 10.46 10.09 10.68 10.71 11.42 10.48 9.633

144 13.25 10.46 10.1 10.7 10.7 11.41 10.52 9.63

145 13.54 10.51 10.1 10.72 10.68 11.39 10.55 9.629

146 13.72 10.5 10.11 10.75 10.66 11.37 10.59 9.633

147 13.91 10.51 10.1 10.77 10.64 11.35 10.63 9.638

148 14.09 10.51 10.1 10.77 10.61 11.33 10.67 9.645

149 14.19 10.55 10.08 10.78 10.59 11.32 10.71 9.655

150 14.35 10.66 10.08 10.8 10.56 11.32 10.74 9.665

151 14.56 10.84 10.07 10.8 10.54 11.32 10.78 9.677

152 14.67 11.1 10.06 10.79 10.52 11.32 10.82 9.688

153 14.85 11.47 10.05 10.79 10.49 11.32 10.85 9.699

154 15.02 11.84 10.05 10.79 10.46 11.32 10.89 9.71

155 15.08 12.12 10.06 10.78 10.44 11.31 10.92 9.722

156 15.31 12.27 10.12 10.77 10.41 11.31 10.96 9.733

157 15.45 12.39 10.22 10.76 10.38 11.3 10.99 9.747

158 15.62 12.5 10.37 10.76 10.34 11.3 11.03 9.761

159 15.69 12.55 10.59 10.77 10.31 11.29 11.07 9.775

160 15.87 12.64 10.95 10.79 10.27 11.28 11.1 9.789

161 16.09 12.71 11.45 10.84 10.22 11.28 11.13 9.802

162 16.27 12.75 11.9 10.92 10.17 11.28 11.16 9.816

163 16.5 12.84 12.16 11.04 10.14 11.28 11.19 9.831

164 16.64 12.93 12.3 11.2 10.11 11.28 11.22 9.846

165 16.85 13.01 12.43 11.41 10.09 11.29 11.26

166 17.02 13.11 12.54 11.69 10.07 11.3 11.28

167 17.23 13.18 12.66 12.01 10.08 11.31 11.31

168 17.45 13.29 12.79 12.33 10.12 11.33 11.33

169 17.67 13.38 12.9 12.6 10.2 11.37 11.36

Chapter 8 Appendix

224

170 17.89 13.46 13.02 12.79 10.31 11.42

171 18.05 13.52 13.13 12.91 10.46 11.5

172 18.34 13.63 13.27 12.99 10.65

173 18.5 13.73 13.39 13.05 10.84

174 18.7 13.85 13.52 13.12

175 19.03 13.99 13.67 13.18

176 19.24 14.1 13.81

177 19.45 14.22

178 19.74

179

180

Documents

Physical, Morphological and Chemical Structure & Property